A Specific Mapping Study Using Fluorescence Sentinel Lymph Node Detection in Patients with Intermediate- and High-risk Prostate Cancer Undergoing Extended Pelvic Lymph Node Dissection

The role of pelvic lymph node dissection (PLND) during radical prostatectomy (RP) remains a matter of continuous debate [1]. Sentinel lymph node (SLN) detection has been advanced as a potential alternative to PLND. In prostate cancer (PCa), the technique was first described using technetium Tc 99m bound to a colloid [2]; however, radioguided SLN detection has not come into widespread use. The use of the fluorescent dye indocyanine green (ICG) may open the door to broader acceptance of SLN techniques in PCa surgery [3]. Against this background, we provided a comprehensive description of lymphatic landing sites per anatomic region of the prostate using this SLN technique. We also evaluated the sensitivity of ICG-based fluorescence SLN detection to detect lymph node metastases in intermediate- and high-risk patients.

Detailed information on patient selection and detection technique is found in Supplement 1. From November 2012 through September 2015, 42 patients presenting with clinically localized intermediate- or high-risk PCa and scheduled for RP gave written informed consent to participate in our prospective study. ICG (Pulsion Medical Systems, Feldkirchen, Germany) was injected transrectally shortly prior to laparotomy. The first 12 patients enrolled received sextant injections (six injections into base, midportion, and apex of each prostatic lobe peripheral zone). The next 30 patients received injections into one of these sites: prostate base bilaterally, midportion bilaterally, apex bilaterally, left lobe (base, midportion, apex), or right lobe (base, midportion, apex). A near-infrared–sensitive probe (Fluobeam; Fluoptics, Grenoble, France) was used to collect fluorescence generated in the tissue under real-time image guidance. Independent of the findings of fluorescence SLN detection, an extended PLND was subsequently performed. An ex vivo fluorescence examination of all dissected lymph nodes was then carried out. Lymphatic landing sites per anatomic region of the prostate were depicted graphically. Diagnostic statistics assessed the value of ICG in detecting lymph node metastases.

Baseline characteristics of the patients are summarized in Supplementary Table 1. All 42 patients had one lymph node or more detected by fluorescence. The lymphatic mapping study showed the following results (Fig. 1; Supplementary Table 2): (1) The external and internal iliac regions encompassed the majority of SLNs; (2) the common iliac regions contained up to 22% of all SLNs; (3) a prostatic lobe can drain into the contralateral group of pelvic lymph nodes; (4) the fossa of Marcille also received significant drainage; and (5) practically all sites of the prostate can drain to different regions of the pelvis bilaterally.

Fig. 1

Percentages of sentinel nodes detected per drainage region with regard to anatomic sites of the prostate. The indocyanine green injection sites are depicted in the upper right corner. The orange zones represent the external iliac regions, the yellow zones represent the internal iliac regions, the green zones represent the common iliac regions, and the regions delineated by dashed lines represent the fossa of Marcille.

 

The drainage pattern did not show that distinct lymphatic pathways exist per prostatic anatomic region. Our results also underscored that lymphatics cross over to the opposite side and that the common iliac regions and the fossa of Marcille should not be overlooked during PLND, as combined, they may contain up to a third of all SLNs. The multitude of lymphatic landing sites as well as the individual variability of lymphatic drainage may represent obstacles to intraoperative SLN detection. In contrast to breast cancer, lymph node metastases in PCa do not follow a predefined pathway of metastatic spread, and there is no certainty that the histologic status of the SLN reflects the status of the entire pelvic node basin.

Among the 12 patients who received systematic ICG injections, a median of 15 SLNs per patient were removed (interquartile range [IQR]: 10–20). Five patients had a total of 29 lymph node metastases (Table 1). Of these, 16 were ICG positive, yielding a sensitivity for the detection of lymph node metastases of 55%. The negative predictive value was 95%, the specificity was 57%, and the predictive positive value was 8%. Repeating analyses with patients as the unit of analysis showed that four of five patients had ICG-positive lymph node metastases, yielding sensitivity of 80% for the correct staging of patients.

Hruby et al evaluated 38 patients who underwent laparoscopic RP [4]. The sensitivity of the SLN technique with ICG to detect metastases was 98% (42 nodes in 15 patients). Yuen et al reported on 66 patients who underwent RP and SNL detection using ICG [5]. Only nine metastases in six patients were detected, all of which were ICG positive. Our median number of nodes removed was 35 versus 18 in both previous reports. These findings suggest that our template leads to a notably more complete PLND, which would result in more precise sensitivity analyses. In addition, our median number of SLNs removed was 15 (IQR: 10–20), which is in the higher range of most previous studies using ICG alone or in combination with a radiocolloid [3], [4], [5], and [6]. Other authors may not have searched as thoroughly for SLNs in surgically less accessible regions such as the proximal part of the common iliac vessels, in the fossa of Marcille, and in the presacral regions. In the study by Yuen et al, only 2% and 1% of all SLNs were found in the common iliac and presacral regions, respectively [5].

The low sensitivity of ICG to detect metastases in the current study raises a concern about “skip” metastases. The presence of tumor cells in the lymph nodes may clog the feeding lymphatics and interfere with their ability to take up ICG, as has been postulated for radiocolloids [7]. Interestingly, we could sometimes visualize lymphatic drainage going around enlarged nodes. Going forward, emerging molecular imaging modalities using dyes conjugated with tumor-specific peptides may provide added sensitivity [8], [9], and [10]. It has to be noted that it is not possible to discriminate with certainty between primary landing sites and higher levels of drainage; however, the fact that fluorescent lymph nodes appeared 15–30 min after ICG injection and that direct drainage through lymphatic vessels could be seen in some cases suggests that these lymph nodes can be considered SLNs. Our study is limited by its small sample size. Nevertheless, it included a high number of lymph node metastases in the setting of extended PLND, allowing appropriate sensitivity analyses.

In conclusion, our lymphatic mapping study delineated a complex drainage pattern of the prostate that raises questions about whether targeted lymph node dissection can be implemented in PCa. Together with the low sensitivity for the detection of metastases, these results suggest that, for the time being, fluorescence SNL detection does not represent an alternative to a meticulously performed PLND in higher risk patients.

Author contributions: George N. Thalmann and Daniel P. Nguyen had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Nguyen, Thalmann.

Acquisition of data: Huber, Metzger, Schudel.

Analysis and interpretation of data: Nguyen, Huber.

Drafting of the manuscript: Nguyen.

Critical revision of the manuscript for important intellectual content: Huber, Metzger, Genitsch, Schudel, Thalmann.

Statistical analysis: Nguyen.

Obtaining funding: None.

Administrative, technical, or material support: None.

Supervision: Nguyen, Thalmann.

Other (specify): None.

Financial disclosures: George N. Thalmann and Daniel P. Nguyen certify that all conflicts of interest, including specific financial interests and relationships and affiliations relevant to the subject matter or materials discussed in the manuscript (eg, employment/ affiliation, grants or funding, consultancies, honoraria, stock ownership or options, expert testimony, royalties, or patents filed, received, or pending), are the following: None.

Funding/Support and role of the sponsor: The Fluobeam imaging device was provided at no charge by Fluoptics (Grenoble, France); however, the authors had complete control of the data and information submitted for publication.

The role of pelvic lymph node dissection (PLND) during radical prostatectomy (RP) remains a matter of continuous debate [1]. Sentinel lymph node (SLN) detection has been advanced as a potential alternative to PLND. In prostate cancer (PCa), the technique was first described using technetium Tc 99m bound to a colloid [2]; however, radioguided SLN detection has not come into widespread use. The use of the fluorescent dye indocyanine green (ICG) may open the door to broader acceptance of SLN techniques in PCa surgery [3]. Against this background, we provided a comprehensive description of lymphatic landing sites per anatomic region of the prostate using this SLN technique. We also evaluated the sensitivity of ICG-based fluorescence SLN detection to detect lymph node metastases in intermediate- and high-risk patients.

Detailed information on patient selection and detection technique is found in Supplement 1. From November 2012 through September 2015, 42 patients presenting with clinically localized intermediate- or high-risk PCa and scheduled for RP gave written informed consent to participate in our prospective study. ICG (Pulsion Medical Systems, Feldkirchen, Germany) was injected transrectally shortly prior to laparotomy. The first 12 patients enrolled received sextant injections (six injections into base, midportion, and apex of each prostatic lobe peripheral zone). The next 30 patients received injections into one of these sites: prostate base bilaterally, midportion bilaterally, apex bilaterally, left lobe (base, midportion, apex), or right lobe (base, midportion, apex). A near-infrared–sensitive probe (Fluobeam; Fluoptics, Grenoble, France) was used to collect fluorescence generated in the tissue under real-time image guidance. Independent of the findings of fluorescence SLN detection, an extended PLND was subsequently performed. An ex vivo fluorescence examination of all dissected lymph nodes was then carried out. Lymphatic landing sites per anatomic region of the prostate were depicted graphically. Diagnostic statistics assessed the value of ICG in detecting lymph node metastases.

Baseline characteristics of the patients are summarized in Supplementary Table 1. All 42 patients had one lymph node or more detected by fluorescence. The lymphatic mapping study showed the following results (Fig. 1; Supplementary Table 2): (1) The external and internal iliac regions encompassed the majority of SLNs; (2) the common iliac regions contained up to 22% of all SLNs; (3) a prostatic lobe can drain into the contralateral group of pelvic lymph nodes; (4) the fossa of Marcille also received significant drainage; and (5) practically all sites of the prostate can drain to different regions of the pelvis bilaterally.

Fig. 1

Percentages of sentinel nodes detected per drainage region with regard to anatomic sites of the prostate. The indocyanine green injection sites are depicted in the upper right corner. The orange zones represent the external iliac regions, the yellow zones represent the internal iliac regions, the green zones represent the common iliac regions, and the regions delineated by dashed lines represent the fossa of Marcille.

 

The drainage pattern did not show that distinct lymphatic pathways exist per prostatic anatomic region. Our results also underscored that lymphatics cross over to the opposite side and that the common iliac regions and the fossa of Marcille should not be overlooked during PLND, as combined, they may contain up to a third of all SLNs. The multitude of lymphatic landing sites as well as the individual variability of lymphatic drainage may represent obstacles to intraoperative SLN detection. In contrast to breast cancer, lymph node metastases in PCa do not follow a predefined pathway of metastatic spread, and there is no certainty that the histologic status of the SLN reflects the status of the entire pelvic node basin.

Among the 12 patients who received systematic ICG injections, a median of 15 SLNs per patient were removed (interquartile range [IQR]: 10–20). Five patients had a total of 29 lymph node metastases (Table 1). Of these, 16 were ICG positive, yielding a sensitivity for the detection of lymph node metastases of 55%. The negative predictive value was 95%, the specificity was 57%, and the predictive positive value was 8%. Repeating analyses with patients as the unit of analysis showed that four of five patients had ICG-positive lymph node metastases, yielding sensitivity of 80% for the correct staging of patients.

Hruby et al evaluated 38 patients who underwent laparoscopic RP [4]. The sensitivity of the SLN technique with ICG to detect metastases was 98% (42 nodes in 15 patients). Yuen et al reported on 66 patients who underwent RP and SNL detection using ICG [5]. Only nine metastases in six patients were detected, all of which were ICG positive. Our median number of nodes removed was 35 versus 18 in both previous reports. These findings suggest that our template leads to a notably more complete PLND, which would result in more precise sensitivity analyses. In addition, our median number of SLNs removed was 15 (IQR: 10–20), which is in the higher range of most previous studies using ICG alone or in combination with a radiocolloid [3], [4], [5], and [6]. Other authors may not have searched as thoroughly for SLNs in surgically less accessible regions such as the proximal part of the common iliac vessels, in the fossa of Marcille, and in the presacral regions. In the study by Yuen et al, only 2% and 1% of all SLNs were found in the common iliac and presacral regions, respectively [5].

The low sensitivity of ICG to detect metastases in the current study raises a concern about “skip” metastases. The presence of tumor cells in the lymph nodes may clog the feeding lymphatics and interfere with their ability to take up ICG, as has been postulated for radiocolloids [7]. Interestingly, we could sometimes visualize lymphatic drainage going around enlarged nodes. Going forward, emerging molecular imaging modalities using dyes conjugated with tumor-specific peptides may provide added sensitivity [8], [9], and [10]. It has to be noted that it is not possible to discriminate with certainty between primary landing sites and higher levels of drainage; however, the fact that fluorescent lymph nodes appeared 15–30 min after ICG injection and that direct drainage through lymphatic vessels could be seen in some cases suggests that these lymph nodes can be considered SLNs. Our study is limited by its small sample size. Nevertheless, it included a high number of lymph node metastases in the setting of extended PLND, allowing appropriate sensitivity analyses.

In conclusion, our lymphatic mapping study delineated a complex drainage pattern of the prostate that raises questions about whether targeted lymph node dissection can be implemented in PCa. Together with the low sensitivity for the detection of metastases, these results suggest that, for the time being, fluorescence SNL detection does not represent an alternative to a meticulously performed PLND in higher risk patients.

Author contributions: George N. Thalmann and Daniel P. Nguyen had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Nguyen, Thalmann.

Acquisition of data: Huber, Metzger, Schudel.

Analysis and interpretation of data: Nguyen, Huber.

Drafting of the manuscript: Nguyen.

Critical revision of the manuscript for important intellectual content: Huber, Metzger, Genitsch, Schudel, Thalmann.

Statistical analysis: Nguyen.

Obtaining funding: None.

Administrative, technical, or material support: None.

Supervision: Nguyen, Thalmann.

Other (specify): None.

Financial disclosures: George N. Thalmann and Daniel P. Nguyen certify that all conflicts of interest, including specific financial interests and relationships and affiliations relevant to the subject matter or materials discussed in the manuscript (eg, employment/ affiliation, grants or funding, consultancies, honoraria, stock ownership or options, expert testimony, royalties, or patents filed, received, or pending), are the following: None.

Funding/Support and role of the sponsor: The Fluobeam imaging device was provided at no charge by Fluoptics (Grenoble, France); however, the authors had complete control of the data and information submitted for publication.

The role of pelvic lymph node dissection (PLND) during radical prostatectomy (RP) remains a matter of continuous debate [1]. Sentinel lymph node (SLN) detection has been advanced as a potential alternative to PLND. In prostate cancer (PCa), the technique was first described using technetium Tc 99m bound to a colloid [2]; however, radioguided SLN detection has not come into widespread use. The use of the fluorescent dye indocyanine green (ICG) may open the door to broader acceptance of SLN techniques in PCa surgery [3]. Against this background, we provided a comprehensive description of lymphatic landing sites per anatomic region of the prostate using this SLN technique. We also evaluated the sensitivity of ICG-based fluorescence SLN detection to detect lymph node metastases in intermediate- and high-risk patients.

Detailed information on patient selection and detection technique is found in Supplement 1. From November 2012 through September 2015, 42 patients presenting with clinically localized intermediate- or high-risk PCa and scheduled for RP gave written informed consent to participate in our prospective study. ICG (Pulsion Medical Systems, Feldkirchen, Germany) was injected transrectally shortly prior to laparotomy. The first 12 patients enrolled received sextant injections (six injections into base, midportion, and apex of each prostatic lobe peripheral zone). The next 30 patients received injections into one of these sites: prostate base bilaterally, midportion bilaterally, apex bilaterally, left lobe (base, midportion, apex), or right lobe (base, midportion, apex). A near-infrared–sensitive probe (Fluobeam; Fluoptics, Grenoble, France) was used to collect fluorescence generated in the tissue under real-time image guidance. Independent of the findings of fluorescence SLN detection, an extended PLND was subsequently performed. An ex vivo fluorescence examination of all dissected lymph nodes was then carried out. Lymphatic landing sites per anatomic region of the prostate were depicted graphically. Diagnostic statistics assessed the value of ICG in detecting lymph node metastases.

Baseline characteristics of the patients are summarized in Supplementary Table 1. All 42 patients had one lymph node or more detected by fluorescence. The lymphatic mapping study showed the following results (Fig. 1; Supplementary Table 2): (1) The external and internal iliac regions encompassed the majority of SLNs; (2) the common iliac regions contained up to 22% of all SLNs; (3) a prostatic lobe can drain into the contralateral group of pelvic lymph nodes; (4) the fossa of Marcille also received significant drainage; and (5) practically all sites of the prostate can drain to different regions of the pelvis bilaterally.

Fig. 1

Percentages of sentinel nodes detected per drainage region with regard to anatomic sites of the prostate. The indocyanine green injection sites are depicted in the upper right corner. The orange zones represent the external iliac regions, the yellow zones represent the internal iliac regions, the green zones represent the common iliac regions, and the regions delineated by dashed lines represent the fossa of Marcille.

 

The drainage pattern did not show that distinct lymphatic pathways exist per prostatic anatomic region. Our results also underscored that lymphatics cross over to the opposite side and that the common iliac regions and the fossa of Marcille should not be overlooked during PLND, as combined, they may contain up to a third of all SLNs. The multitude of lymphatic landing sites as well as the individual variability of lymphatic drainage may represent obstacles to intraoperative SLN detection. In contrast to breast cancer, lymph node metastases in PCa do not follow a predefined pathway of metastatic spread, and there is no certainty that the histologic status of the SLN reflects the status of the entire pelvic node basin.

Among the 12 patients who received systematic ICG injections, a median of 15 SLNs per patient were removed (interquartile range [IQR]: 10–20). Five patients had a total of 29 lymph node metastases (Table 1). Of these, 16 were ICG positive, yielding a sensitivity for the detection of lymph node metastases of 55%. The negative predictive value was 95%, the specificity was 57%, and the predictive positive value was 8%. Repeating analyses with patients as the unit of analysis showed that four of five patients had ICG-positive lymph node metastases, yielding sensitivity of 80% for the correct staging of patients.

Hruby et al evaluated 38 patients who underwent laparoscopic RP [4]. The sensitivity of the SLN technique with ICG to detect metastases was 98% (42 nodes in 15 patients). Yuen et al reported on 66 patients who underwent RP and SNL detection using ICG [5]. Only nine metastases in six patients were detected, all of which were ICG positive. Our median number of nodes removed was 35 versus 18 in both previous reports. These findings suggest that our template leads to a notably more complete PLND, which would result in more precise sensitivity analyses. In addition, our median number of SLNs removed was 15 (IQR: 10–20), which is in the higher range of most previous studies using ICG alone or in combination with a radiocolloid [3], [4], [5], and [6]. Other authors may not have searched as thoroughly for SLNs in surgically less accessible regions such as the proximal part of the common iliac vessels, in the fossa of Marcille, and in the presacral regions. In the study by Yuen et al, only 2% and 1% of all SLNs were found in the common iliac and presacral regions, respectively [5].

The low sensitivity of ICG to detect metastases in the current study raises a concern about “skip” metastases. The presence of tumor cells in the lymph nodes may clog the feeding lymphatics and interfere with their ability to take up ICG, as has been postulated for radiocolloids [7]. Interestingly, we could sometimes visualize lymphatic drainage going around enlarged nodes. Going forward, emerging molecular imaging modalities using dyes conjugated with tumor-specific peptides may provide added sensitivity [8], [9], and [10]. It has to be noted that it is not possible to discriminate with certainty between primary landing sites and higher levels of drainage; however, the fact that fluorescent lymph nodes appeared 15–30 min after ICG injection and that direct drainage through lymphatic vessels could be seen in some cases suggests that these lymph nodes can be considered SLNs. Our study is limited by its small sample size. Nevertheless, it included a high number of lymph node metastases in the setting of extended PLND, allowing appropriate sensitivity analyses.

In conclusion, our lymphatic mapping study delineated a complex drainage pattern of the prostate that raises questions about whether targeted lymph node dissection can be implemented in PCa. Together with the low sensitivity for the detection of metastases, these results suggest that, for the time being, fluorescence SNL detection does not represent an alternative to a meticulously performed PLND in higher risk patients.

Author contributions: George N. Thalmann and Daniel P. Nguyen had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Nguyen, Thalmann.

Acquisition of data: Huber, Metzger, Schudel.

Analysis and interpretation of data: Nguyen, Huber.

Drafting of the manuscript: Nguyen.

Critical revision of the manuscript for important intellectual content: Huber, Metzger, Genitsch, Schudel, Thalmann.

Statistical analysis: Nguyen.

Obtaining funding: None.

Administrative, technical, or material support: None.

Supervision: Nguyen, Thalmann.

Other (specify): None.

Financial disclosures: George N. Thalmann and Daniel P. Nguyen certify that all conflicts of interest, including specific financial interests and relationships and affiliations relevant to the subject matter or materials discussed in the manuscript (eg, employment/ affiliation, grants or funding, consultancies, honoraria, stock ownership or options, expert testimony, royalties, or patents filed, received, or pending), are the following: None.

Funding/Support and role of the sponsor: The Fluobeam imaging device was provided at no charge by Fluoptics (Grenoble, France); however, the authors had complete control of the data and information submitted for publication.

The role of pelvic lymph node dissection (PLND) during radical prostatectomy (RP) remains a matter of continuous debate [1]. Sentinel lymph node (SLN) detection has been advanced as a potential alternative to PLND. In prostate cancer (PCa), the technique was first described using technetium Tc 99m bound to a colloid [2]; however, radioguided SLN detection has not come into widespread use. The use of the fluorescent dye indocyanine green (ICG) may open the door to broader acceptance of SLN techniques in PCa surgery [3]. Against this background, we provided a comprehensive description of lymphatic landing sites per anatomic region of the prostate using this SLN technique. We also evaluated the sensitivity of ICG-based fluorescence SLN detection to detect lymph node metastases in intermediate- and high-risk patients.

Detailed information on patient selection and detection technique is found in Supplement 1. From November 2012 through September 2015, 42 patients presenting with clinically localized intermediate- or high-risk PCa and scheduled for RP gave written informed consent to participate in our prospective study. ICG (Pulsion Medical Systems, Feldkirchen, Germany) was injected transrectally shortly prior to laparotomy. The first 12 patients enrolled received sextant injections (six injections into base, midportion, and apex of each prostatic lobe peripheral zone). The next 30 patients received injections into one of these sites: prostate base bilaterally, midportion bilaterally, apex bilaterally, left lobe (base, midportion, apex), or right lobe (base, midportion, apex). A near-infrared–sensitive probe (Fluobeam; Fluoptics, Grenoble, France) was used to collect fluorescence generated in the tissue under real-time image guidance. Independent of the findings of fluorescence SLN detection, an extended PLND was subsequently performed. An ex vivo fluorescence examination of all dissected lymph nodes was then carried out. Lymphatic landing sites per anatomic region of the prostate were depicted graphically. Diagnostic statistics assessed the value of ICG in detecting lymph node metastases.

Baseline characteristics of the patients are summarized in Supplementary Table 1. All 42 patients had one lymph node or more detected by fluorescence. The lymphatic mapping study showed the following results (Fig. 1; Supplementary Table 2): (1) The external and internal iliac regions encompassed the majority of SLNs; (2) the common iliac regions contained up to 22% of all SLNs; (3) a prostatic lobe can drain into the contralateral group of pelvic lymph nodes; (4) the fossa of Marcille also received significant drainage; and (5) practically all sites of the prostate can drain to different regions of the pelvis bilaterally.

Fig. 1

Percentages of sentinel nodes detected per drainage region with regard to anatomic sites of the prostate. The indocyanine green injection sites are depicted in the upper right corner. The orange zones represent the external iliac regions, the yellow zones represent the internal iliac regions, the green zones represent the common iliac regions, and the regions delineated by dashed lines represent the fossa of Marcille.

 

The drainage pattern did not show that distinct lymphatic pathways exist per prostatic anatomic region. Our results also underscored that lymphatics cross over to the opposite side and that the common iliac regions and the fossa of Marcille should not be overlooked during PLND, as combined, they may contain up to a third of all SLNs. The multitude of lymphatic landing sites as well as the individual variability of lymphatic drainage may represent obstacles to intraoperative SLN detection. In contrast to breast cancer, lymph node metastases in PCa do not follow a predefined pathway of metastatic spread, and there is no certainty that the histologic status of the SLN reflects the status of the entire pelvic node basin.

Among the 12 patients who received systematic ICG injections, a median of 15 SLNs per patient were removed (interquartile range [IQR]: 10–20). Five patients had a total of 29 lymph node metastases (Table 1). Of these, 16 were ICG positive, yielding a sensitivity for the detection of lymph node metastases of 55%. The negative predictive value was 95%, the specificity was 57%, and the predictive positive value was 8%. Repeating analyses with patients as the unit of analysis showed that four of five patients had ICG-positive lymph node metastases, yielding sensitivity of 80% for the correct staging of patients.

Hruby et al evaluated 38 patients who underwent laparoscopic RP [4]. The sensitivity of the SLN technique with ICG to detect metastases was 98% (42 nodes in 15 patients). Yuen et al reported on 66 patients who underwent RP and SNL detection using ICG [5]. Only nine metastases in six patients were detected, all of which were ICG positive. Our median number of nodes removed was 35 versus 18 in both previous reports. These findings suggest that our template leads to a notably more complete PLND, which would result in more precise sensitivity analyses. In addition, our median number of SLNs removed was 15 (IQR: 10–20), which is in the higher range of most previous studies using ICG alone or in combination with a radiocolloid [3], [4], [5], and [6]. Other authors may not have searched as thoroughly for SLNs in surgically less accessible regions such as the proximal part of the common iliac vessels, in the fossa of Marcille, and in the presacral regions. In the study by Yuen et al, only 2% and 1% of all SLNs were found in the common iliac and presacral regions, respectively [5].

The low sensitivity of ICG to detect metastases in the current study raises a concern about “skip” metastases. The presence of tumor cells in the lymph nodes may clog the feeding lymphatics and interfere with their ability to take up ICG, as has been postulated for radiocolloids [7]. Interestingly, we could sometimes visualize lymphatic drainage going around enlarged nodes. Going forward, emerging molecular imaging modalities using dyes conjugated with tumor-specific peptides may provide added sensitivity [8], [9], and [10]. It has to be noted that it is not possible to discriminate with certainty between primary landing sites and higher levels of drainage; however, the fact that fluorescent lymph nodes appeared 15–30 min after ICG injection and that direct drainage through lymphatic vessels could be seen in some cases suggests that these lymph nodes can be considered SLNs. Our study is limited by its small sample size. Nevertheless, it included a high number of lymph node metastases in the setting of extended PLND, allowing appropriate sensitivity analyses.

In conclusion, our lymphatic mapping study delineated a complex drainage pattern of the prostate that raises questions about whether targeted lymph node dissection can be implemented in PCa. Together with the low sensitivity for the detection of metastases, these results suggest that, for the time being, fluorescence SNL detection does not represent an alternative to a meticulously performed PLND in higher risk patients.

Author contributions: George N. Thalmann and Daniel P. Nguyen had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Nguyen, Thalmann.

Acquisition of data: Huber, Metzger, Schudel.

Analysis and interpretation of data: Nguyen, Huber.

Drafting of the manuscript: Nguyen.

Critical revision of the manuscript for important intellectual content: Huber, Metzger, Genitsch, Schudel, Thalmann.

Statistical analysis: Nguyen.

Obtaining funding: None.

Administrative, technical, or material support: None.

Supervision: Nguyen, Thalmann.

Other (specify): None.

Financial disclosures: George N. Thalmann and Daniel P. Nguyen certify that all conflicts of interest, including specific financial interests and relationships and affiliations relevant to the subject matter or materials discussed in the manuscript (eg, employment/ affiliation, grants or funding, consultancies, honoraria, stock ownership or options, expert testimony, royalties, or patents filed, received, or pending), are the following: None.

Funding/Support and role of the sponsor: The Fluobeam imaging device was provided at no charge by Fluoptics (Grenoble, France); however, the authors had complete control of the data and information submitted for publication.

The role of pelvic lymph node dissection (PLND) during radical prostatectomy (RP) remains a matter of continuous debate [1]. Sentinel lymph node (SLN) detection has been advanced as a potential alternative to PLND. In prostate cancer (PCa), the technique was first described using technetium Tc 99m bound to a colloid [2]; however, radioguided SLN detection has not come into widespread use. The use of the fluorescent dye indocyanine green (ICG) may open the door to broader acceptance of SLN techniques in PCa surgery [3]. Against this background, we provided a comprehensive description of lymphatic landing sites per anatomic region of the prostate using this SLN technique. We also evaluated the sensitivity of ICG-based fluorescence SLN detection to detect lymph node metastases in intermediate- and high-risk patients.

Detailed information on patient selection and detection technique is found in Supplement 1. From November 2012 through September 2015, 42 patients presenting with clinically localized intermediate- or high-risk PCa and scheduled for RP gave written informed consent to participate in our prospective study. ICG (Pulsion Medical Systems, Feldkirchen, Germany) was injected transrectally shortly prior to laparotomy. The first 12 patients enrolled received sextant injections (six injections into base, midportion, and apex of each prostatic lobe peripheral zone). The next 30 patients received injections into one of these sites: prostate base bilaterally, midportion bilaterally, apex bilaterally, left lobe (base, midportion, apex), or right lobe (base, midportion, apex). A near-infrared–sensitive probe (Fluobeam; Fluoptics, Grenoble, France) was used to collect fluorescence generated in the tissue under real-time image guidance. Independent of the findings of fluorescence SLN detection, an extended PLND was subsequently performed. An ex vivo fluorescence examination of all dissected lymph nodes was then carried out. Lymphatic landing sites per anatomic region of the prostate were depicted graphically. Diagnostic statistics assessed the value of ICG in detecting lymph node metastases.

Baseline characteristics of the patients are summarized in Supplementary Table 1. All 42 patients had one lymph node or more detected by fluorescence. The lymphatic mapping study showed the following results (Fig. 1; Supplementary Table 2): (1) The external and internal iliac regions encompassed the majority of SLNs; (2) the common iliac regions contained up to 22% of all SLNs; (3) a prostatic lobe can drain into the contralateral group of pelvic lymph nodes; (4) the fossa of Marcille also received significant drainage; and (5) practically all sites of the prostate can drain to different regions of the pelvis bilaterally.

Fig. 1

Percentages of sentinel nodes detected per drainage region with regard to anatomic sites of the prostate. The indocyanine green injection sites are depicted in the upper right corner. The orange zones represent the external iliac regions, the yellow zones represent the internal iliac regions, the green zones represent the common iliac regions, and the regions delineated by dashed lines represent the fossa of Marcille.

 

The drainage pattern did not show that distinct lymphatic pathways exist per prostatic anatomic region. Our results also underscored that lymphatics cross over to the opposite side and that the common iliac regions and the fossa of Marcille should not be overlooked during PLND, as combined, they may contain up to a third of all SLNs. The multitude of lymphatic landing sites as well as the individual variability of lymphatic drainage may represent obstacles to intraoperative SLN detection. In contrast to breast cancer, lymph node metastases in PCa do not follow a predefined pathway of metastatic spread, and there is no certainty that the histologic status of the SLN reflects the status of the entire pelvic node basin.

Among the 12 patients who received systematic ICG injections, a median of 15 SLNs per patient were removed (interquartile range [IQR]: 10–20). Five patients had a total of 29 lymph node metastases (Table 1). Of these, 16 were ICG positive, yielding a sensitivity for the detection of lymph node metastases of 55%. The negative predictive value was 95%, the specificity was 57%, and the predictive positive value was 8%. Repeating analyses with patients as the unit of analysis showed that four of five patients had ICG-positive lymph node metastases, yielding sensitivity of 80% for the correct staging of patients.

Hruby et al evaluated 38 patients who underwent laparoscopic RP [4]. The sensitivity of the SLN technique with ICG to detect metastases was 98% (42 nodes in 15 patients). Yuen et al reported on 66 patients who underwent RP and SNL detection using ICG [5]. Only nine metastases in six patients were detected, all of which were ICG positive. Our median number of nodes removed was 35 versus 18 in both previous reports. These findings suggest that our template leads to a notably more complete PLND, which would result in more precise sensitivity analyses. In addition, our median number of SLNs removed was 15 (IQR: 10–20), which is in the higher range of most previous studies using ICG alone or in combination with a radiocolloid [3], [4], [5], and [6]. Other authors may not have searched as thoroughly for SLNs in surgically less accessible regions such as the proximal part of the common iliac vessels, in the fossa of Marcille, and in the presacral regions. In the study by Yuen et al, only 2% and 1% of all SLNs were found in the common iliac and presacral regions, respectively [5].

The low sensitivity of ICG to detect metastases in the current study raises a concern about “skip” metastases. The presence of tumor cells in the lymph nodes may clog the feeding lymphatics and interfere with their ability to take up ICG, as has been postulated for radiocolloids [7]. Interestingly, we could sometimes visualize lymphatic drainage going around enlarged nodes. Going forward, emerging molecular imaging modalities using dyes conjugated with tumor-specific peptides may provide added sensitivity [8], [9], and [10]. It has to be noted that it is not possible to discriminate with certainty between primary landing sites and higher levels of drainage; however, the fact that fluorescent lymph nodes appeared 15–30 min after ICG injection and that direct drainage through lymphatic vessels could be seen in some cases suggests that these lymph nodes can be considered SLNs. Our study is limited by its small sample size. Nevertheless, it included a high number of lymph node metastases in the setting of extended PLND, allowing appropriate sensitivity analyses.

In conclusion, our lymphatic mapping study delineated a complex drainage pattern of the prostate that raises questions about whether targeted lymph node dissection can be implemented in PCa. Together with the low sensitivity for the detection of metastases, these results suggest that, for the time being, fluorescence SNL detection does not represent an alternative to a meticulously performed PLND in higher risk patients.

Author contributions: George N. Thalmann and Daniel P. Nguyen had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Nguyen, Thalmann.

Acquisition of data: Huber, Metzger, Schudel.

Analysis and interpretation of data: Nguyen, Huber.

Drafting of the manuscript: Nguyen.

Critical revision of the manuscript for important intellectual content: Huber, Metzger, Genitsch, Schudel, Thalmann.

Statistical analysis: Nguyen.

Obtaining funding: None.

Administrative, technical, or material support: None.

Supervision: Nguyen, Thalmann.

Other (specify): None.

Financial disclosures: George N. Thalmann and Daniel P. Nguyen certify that all conflicts of interest, including specific financial interests and relationships and affiliations relevant to the subject matter or materials discussed in the manuscript (eg, employment/ affiliation, grants or funding, consultancies, honoraria, stock ownership or options, expert testimony, royalties, or patents filed, received, or pending), are the following: None.

Funding/Support and role of the sponsor: The Fluobeam imaging device was provided at no charge by Fluoptics (Grenoble, France); however, the authors had complete control of the data and information submitted for publication.

The role of pelvic lymph node dissection (PLND) during radical prostatectomy (RP) remains a matter of continuous debate [1]. Sentinel lymph node (SLN) detection has been advanced as a potential alternative to PLND. In prostate cancer (PCa), the technique was first described using technetium Tc 99m bound to a colloid [2]; however, radioguided SLN detection has not come into widespread use. The use of the fluorescent dye indocyanine green (ICG) may open the door to broader acceptance of SLN techniques in PCa surgery [3]. Against this background, we provided a comprehensive description of lymphatic landing sites per anatomic region of the prostate using this SLN technique. We also evaluated the sensitivity of ICG-based fluorescence SLN detection to detect lymph node metastases in intermediate- and high-risk patients.

Detailed information on patient selection and detection technique is found in Supplement 1. From November 2012 through September 2015, 42 patients presenting with clinically localized intermediate- or high-risk PCa and scheduled for RP gave written informed consent to participate in our prospective study. ICG (Pulsion Medical Systems, Feldkirchen, Germany) was injected transrectally shortly prior to laparotomy. The first 12 patients enrolled received sextant injections (six injections into base, midportion, and apex of each prostatic lobe peripheral zone). The next 30 patients received injections into one of these sites: prostate base bilaterally, midportion bilaterally, apex bilaterally, left lobe (base, midportion, apex), or right lobe (base, midportion, apex). A near-infrared–sensitive probe (Fluobeam; Fluoptics, Grenoble, France) was used to collect fluorescence generated in the tissue under real-time image guidance. Independent of the findings of fluorescence SLN detection, an extended PLND was subsequently performed. An ex vivo fluorescence examination of all dissected lymph nodes was then carried out. Lymphatic landing sites per anatomic region of the prostate were depicted graphically. Diagnostic statistics assessed the value of ICG in detecting lymph node metastases.

Baseline characteristics of the patients are summarized in Supplementary Table 1. All 42 patients had one lymph node or more detected by fluorescence. The lymphatic mapping study showed the following results (Fig. 1; Supplementary Table 2): (1) The external and internal iliac regions encompassed the majority of SLNs; (2) the common iliac regions contained up to 22% of all SLNs; (3) a prostatic lobe can drain into the contralateral group of pelvic lymph nodes; (4) the fossa of Marcille also received significant drainage; and (5) practically all sites of the prostate can drain to different regions of the pelvis bilaterally.

Fig. 1

Percentages of sentinel nodes detected per drainage region with regard to anatomic sites of the prostate. The indocyanine green injection sites are depicted in the upper right corner. The orange zones represent the external iliac regions, the yellow zones represent the internal iliac regions, the green zones represent the common iliac regions, and the regions delineated by dashed lines represent the fossa of Marcille.

 

The drainage pattern did not show that distinct lymphatic pathways exist per prostatic anatomic region. Our results also underscored that lymphatics cross over to the opposite side and that the common iliac regions and the fossa of Marcille should not be overlooked during PLND, as combined, they may contain up to a third of all SLNs. The multitude of lymphatic landing sites as well as the individual variability of lymphatic drainage may represent obstacles to intraoperative SLN detection. In contrast to breast cancer, lymph node metastases in PCa do not follow a predefined pathway of metastatic spread, and there is no certainty that the histologic status of the SLN reflects the status of the entire pelvic node basin.

Among the 12 patients who received systematic ICG injections, a median of 15 SLNs per patient were removed (interquartile range [IQR]: 10–20). Five patients had a total of 29 lymph node metastases (Table 1). Of these, 16 were ICG positive, yielding a sensitivity for the detection of lymph node metastases of 55%. The negative predictive value was 95%, the specificity was 57%, and the predictive positive value was 8%. Repeating analyses with patients as the unit of analysis showed that four of five patients had ICG-positive lymph node metastases, yielding sensitivity of 80% for the correct staging of patients.

Hruby et al evaluated 38 patients who underwent laparoscopic RP [4]. The sensitivity of the SLN technique with ICG to detect metastases was 98% (42 nodes in 15 patients). Yuen et al reported on 66 patients who underwent RP and SNL detection using ICG [5]. Only nine metastases in six patients were detected, all of which were ICG positive. Our median number of nodes removed was 35 versus 18 in both previous reports. These findings suggest that our template leads to a notably more complete PLND, which would result in more precise sensitivity analyses. In addition, our median number of SLNs removed was 15 (IQR: 10–20), which is in the higher range of most previous studies using ICG alone or in combination with a radiocolloid [3], [4], [5], and [6]. Other authors may not have searched as thoroughly for SLNs in surgically less accessible regions such as the proximal part of the common iliac vessels, in the fossa of Marcille, and in the presacral regions. In the study by Yuen et al, only 2% and 1% of all SLNs were found in the common iliac and presacral regions, respectively [5].

The low sensitivity of ICG to detect metastases in the current study raises a concern about “skip” metastases. The presence of tumor cells in the lymph nodes may clog the feeding lymphatics and interfere with their ability to take up ICG, as has been postulated for radiocolloids [7]. Interestingly, we could sometimes visualize lymphatic drainage going around enlarged nodes. Going forward, emerging molecular imaging modalities using dyes conjugated with tumor-specific peptides may provide added sensitivity [8], [9], and [10]. It has to be noted that it is not possible to discriminate with certainty between primary landing sites and higher levels of drainage; however, the fact that fluorescent lymph nodes appeared 15–30 min after ICG injection and that direct drainage through lymphatic vessels could be seen in some cases suggests that these lymph nodes can be considered SLNs. Our study is limited by its small sample size. Nevertheless, it included a high number of lymph node metastases in the setting of extended PLND, allowing appropriate sensitivity analyses.

In conclusion, our lymphatic mapping study delineated a complex drainage pattern of the prostate that raises questions about whether targeted lymph node dissection can be implemented in PCa. Together with the low sensitivity for the detection of metastases, these results suggest that, for the time being, fluorescence SNL detection does not represent an alternative to a meticulously performed PLND in higher risk patients.

Author contributions: George N. Thalmann and Daniel P. Nguyen had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Nguyen, Thalmann.

Acquisition of data: Huber, Metzger, Schudel.

Analysis and interpretation of data: Nguyen, Huber.

Drafting of the manuscript: Nguyen.

Critical revision of the manuscript for important intellectual content: Huber, Metzger, Genitsch, Schudel, Thalmann.

Statistical analysis: Nguyen.

Obtaining funding: None.

Administrative, technical, or material support: None.

Supervision: Nguyen, Thalmann.

Other (specify): None.

Financial disclosures: George N. Thalmann and Daniel P. Nguyen certify that all conflicts of interest, including specific financial interests and relationships and affiliations relevant to the subject matter or materials discussed in the manuscript (eg, employment/ affiliation, grants or funding, consultancies, honoraria, stock ownership or options, expert testimony, royalties, or patents filed, received, or pending), are the following: None.

Funding/Support and role of the sponsor: The Fluobeam imaging device was provided at no charge by Fluoptics (Grenoble, France); however, the authors had complete control of the data and information submitted for publication.

The role of pelvic lymph node dissection (PLND) during radical prostatectomy (RP) remains a matter of continuous debate [1]. Sentinel lymph node (SLN) detection has been advanced as a potential alternative to PLND. In prostate cancer (PCa), the technique was first described using technetium Tc 99m bound to a colloid [2]; however, radioguided SLN detection has not come into widespread use. The use of the fluorescent dye indocyanine green (ICG) may open the door to broader acceptance of SLN techniques in PCa surgery [3]. Against this background, we provided a comprehensive description of lymphatic landing sites per anatomic region of the prostate using this SLN technique. We also evaluated the sensitivity of ICG-based fluorescence SLN detection to detect lymph node metastases in intermediate- and high-risk patients.

Detailed information on patient selection and detection technique is found in Supplement 1. From November 2012 through September 2015, 42 patients presenting with clinically localized intermediate- or high-risk PCa and scheduled for RP gave written informed consent to participate in our prospective study. ICG (Pulsion Medical Systems, Feldkirchen, Germany) was injected transrectally shortly prior to laparotomy. The first 12 patients enrolled received sextant injections (six injections into base, midportion, and apex of each prostatic lobe peripheral zone). The next 30 patients received injections into one of these sites: prostate base bilaterally, midportion bilaterally, apex bilaterally, left lobe (base, midportion, apex), or right lobe (base, midportion, apex). A near-infrared–sensitive probe (Fluobeam; Fluoptics, Grenoble, France) was used to collect fluorescence generated in the tissue under real-time image guidance. Independent of the findings of fluorescence SLN detection, an extended PLND was subsequently performed. An ex vivo fluorescence examination of all dissected lymph nodes was then carried out. Lymphatic landing sites per anatomic region of the prostate were depicted graphically. Diagnostic statistics assessed the value of ICG in detecting lymph node metastases.

Baseline characteristics of the patients are summarized in Supplementary Table 1. All 42 patients had one lymph node or more detected by fluorescence. The lymphatic mapping study showed the following results (Fig. 1; Supplementary Table 2): (1) The external and internal iliac regions encompassed the majority of SLNs; (2) the common iliac regions contained up to 22% of all SLNs; (3) a prostatic lobe can drain into the contralateral group of pelvic lymph nodes; (4) the fossa of Marcille also received significant drainage; and (5) practically all sites of the prostate can drain to different regions of the pelvis bilaterally.

Fig. 1

Percentages of sentinel nodes detected per drainage region with regard to anatomic sites of the prostate. The indocyanine green injection sites are depicted in the upper right corner. The orange zones represent the external iliac regions, the yellow zones represent the internal iliac regions, the green zones represent the common iliac regions, and the regions delineated by dashed lines represent the fossa of Marcille.

 

The drainage pattern did not show that distinct lymphatic pathways exist per prostatic anatomic region. Our results also underscored that lymphatics cross over to the opposite side and that the common iliac regions and the fossa of Marcille should not be overlooked during PLND, as combined, they may contain up to a third of all SLNs. The multitude of lymphatic landing sites as well as the individual variability of lymphatic drainage may represent obstacles to intraoperative SLN detection. In contrast to breast cancer, lymph node metastases in PCa do not follow a predefined pathway of metastatic spread, and there is no certainty that the histologic status of the SLN reflects the status of the entire pelvic node basin.

Among the 12 patients who received systematic ICG injections, a median of 15 SLNs per patient were removed (interquartile range [IQR]: 10–20). Five patients had a total of 29 lymph node metastases (Table 1). Of these, 16 were ICG positive, yielding a sensitivity for the detection of lymph node metastases of 55%. The negative predictive value was 95%, the specificity was 57%, and the predictive positive value was 8%. Repeating analyses with patients as the unit of analysis showed that four of five patients had ICG-positive lymph node metastases, yielding sensitivity of 80% for the correct staging of patients.

Hruby et al evaluated 38 patients who underwent laparoscopic RP [4]. The sensitivity of the SLN technique with ICG to detect metastases was 98% (42 nodes in 15 patients). Yuen et al reported on 66 patients who underwent RP and SNL detection using ICG [5]. Only nine metastases in six patients were detected, all of which were ICG positive. Our median number of nodes removed was 35 versus 18 in both previous reports. These findings suggest that our template leads to a notably more complete PLND, which would result in more precise sensitivity analyses. In addition, our median number of SLNs removed was 15 (IQR: 10–20), which is in the higher range of most previous studies using ICG alone or in combination with a radiocolloid [3], [4], [5], and [6]. Other authors may not have searched as thoroughly for SLNs in surgically less accessible regions such as the proximal part of the common iliac vessels, in the fossa of Marcille, and in the presacral regions. In the study by Yuen et al, only 2% and 1% of all SLNs were found in the common iliac and presacral regions, respectively [5].

The low sensitivity of ICG to detect metastases in the current study raises a concern about “skip” metastases. The presence of tumor cells in the lymph nodes may clog the feeding lymphatics and interfere with their ability to take up ICG, as has been postulated for radiocolloids [7]. Interestingly, we could sometimes visualize lymphatic drainage going around enlarged nodes. Going forward, emerging molecular imaging modalities using dyes conjugated with tumor-specific peptides may provide added sensitivity [8], [9], and [10]. It has to be noted that it is not possible to discriminate with certainty between primary landing sites and higher levels of drainage; however, the fact that fluorescent lymph nodes appeared 15–30 min after ICG injection and that direct drainage through lymphatic vessels could be seen in some cases suggests that these lymph nodes can be considered SLNs. Our study is limited by its small sample size. Nevertheless, it included a high number of lymph node metastases in the setting of extended PLND, allowing appropriate sensitivity analyses.

In conclusion, our lymphatic mapping study delineated a complex drainage pattern of the prostate that raises questions about whether targeted lymph node dissection can be implemented in PCa. Together with the low sensitivity for the detection of metastases, these results suggest that, for the time being, fluorescence SNL detection does not represent an alternative to a meticulously performed PLND in higher risk patients.

Author contributions: George N. Thalmann and Daniel P. Nguyen had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Nguyen, Thalmann.

Acquisition of data: Huber, Metzger, Schudel.

Analysis and interpretation of data: Nguyen, Huber.

Drafting of the manuscript: Nguyen.

Critical revision of the manuscript for important intellectual content: Huber, Metzger, Genitsch, Schudel, Thalmann.

Statistical analysis: Nguyen.

Obtaining funding: None.

Administrative, technical, or material support: None.

Supervision: Nguyen, Thalmann.

Other (specify): None.

Financial disclosures: George N. Thalmann and Daniel P. Nguyen certify that all conflicts of interest, including specific financial interests and relationships and affiliations relevant to the subject matter or materials discussed in the manuscript (eg, employment/ affiliation, grants or funding, consultancies, honoraria, stock ownership or options, expert testimony, royalties, or patents filed, received, or pending), are the following: None.

Funding/Support and role of the sponsor: The Fluobeam imaging device was provided at no charge by Fluoptics (Grenoble, France); however, the authors had complete control of the data and information submitted for publication.

The role of pelvic lymph node dissection (PLND) during radical prostatectomy (RP) remains a matter of continuous debate [1]. Sentinel lymph node (SLN) detection has been advanced as a potential alternative to PLND. In prostate cancer (PCa), the technique was first described using technetium Tc 99m bound to a colloid [2]; however, radioguided SLN detection has not come into widespread use. The use of the fluorescent dye indocyanine green (ICG) may open the door to broader acceptance of SLN techniques in PCa surgery [3]. Against this background, we provided a comprehensive description of lymphatic landing sites per anatomic region of the prostate using this SLN technique. We also evaluated the sensitivity of ICG-based fluorescence SLN detection to detect lymph node metastases in intermediate- and high-risk patients.

Detailed information on patient selection and detection technique is found in Supplement 1. From November 2012 through September 2015, 42 patients presenting with clinically localized intermediate- or high-risk PCa and scheduled for RP gave written informed consent to participate in our prospective study. ICG (Pulsion Medical Systems, Feldkirchen, Germany) was injected transrectally shortly prior to laparotomy. The first 12 patients enrolled received sextant injections (six injections into base, midportion, and apex of each prostatic lobe peripheral zone). The next 30 patients received injections into one of these sites: prostate base bilaterally, midportion bilaterally, apex bilaterally, left lobe (base, midportion, apex), or right lobe (base, midportion, apex). A near-infrared–sensitive probe (Fluobeam; Fluoptics, Grenoble, France) was used to collect fluorescence generated in the tissue under real-time image guidance. Independent of the findings of fluorescence SLN detection, an extended PLND was subsequently performed. An ex vivo fluorescence examination of all dissected lymph nodes was then carried out. Lymphatic landing sites per anatomic region of the prostate were depicted graphically. Diagnostic statistics assessed the value of ICG in detecting lymph node metastases.

Baseline characteristics of the patients are summarized in Supplementary Table 1. All 42 patients had one lymph node or more detected by fluorescence. The lymphatic mapping study showed the following results (Fig. 1; Supplementary Table 2): (1) The external and internal iliac regions encompassed the majority of SLNs; (2) the common iliac regions contained up to 22% of all SLNs; (3) a prostatic lobe can drain into the contralateral group of pelvic lymph nodes; (4) the fossa of Marcille also received significant drainage; and (5) practically all sites of the prostate can drain to different regions of the pelvis bilaterally.

Fig. 1

Percentages of sentinel nodes detected per drainage region with regard to anatomic sites of the prostate. The indocyanine green injection sites are depicted in the upper right corner. The orange zones represent the external iliac regions, the yellow zones represent the internal iliac regions, the green zones represent the common iliac regions, and the regions delineated by dashed lines represent the fossa of Marcille.

 

The drainage pattern did not show that distinct lymphatic pathways exist per prostatic anatomic region. Our results also underscored that lymphatics cross over to the opposite side and that the common iliac regions and the fossa of Marcille should not be overlooked during PLND, as combined, they may contain up to a third of all SLNs. The multitude of lymphatic landing sites as well as the individual variability of lymphatic drainage may represent obstacles to intraoperative SLN detection. In contrast to breast cancer, lymph node metastases in PCa do not follow a predefined pathway of metastatic spread, and there is no certainty that the histologic status of the SLN reflects the status of the entire pelvic node basin.

Among the 12 patients who received systematic ICG injections, a median of 15 SLNs per patient were removed (interquartile range [IQR]: 10–20). Five patients had a total of 29 lymph node metastases (Table 1). Of these, 16 were ICG positive, yielding a sensitivity for the detection of lymph node metastases of 55%. The negative predictive value was 95%, the specificity was 57%, and the predictive positive value was 8%. Repeating analyses with patients as the unit of analysis showed that four of five patients had ICG-positive lymph node metastases, yielding sensitivity of 80% for the correct staging of patients.

Hruby et al evaluated 38 patients who underwent laparoscopic RP [4]. The sensitivity of the SLN technique with ICG to detect metastases was 98% (42 nodes in 15 patients). Yuen et al reported on 66 patients who underwent RP and SNL detection using ICG [5]. Only nine metastases in six patients were detected, all of which were ICG positive. Our median number of nodes removed was 35 versus 18 in both previous reports. These findings suggest that our template leads to a notably more complete PLND, which would result in more precise sensitivity analyses. In addition, our median number of SLNs removed was 15 (IQR: 10–20), which is in the higher range of most previous studies using ICG alone or in combination with a radiocolloid [3], [4], [5], and [6]. Other authors may not have searched as thoroughly for SLNs in surgically less accessible regions such as the proximal part of the common iliac vessels, in the fossa of Marcille, and in the presacral regions. In the study by Yuen et al, only 2% and 1% of all SLNs were found in the common iliac and presacral regions, respectively [5].

The low sensitivity of ICG to detect metastases in the current study raises a concern about “skip” metastases. The presence of tumor cells in the lymph nodes may clog the feeding lymphatics and interfere with their ability to take up ICG, as has been postulated for radiocolloids [7]. Interestingly, we could sometimes visualize lymphatic drainage going around enlarged nodes. Going forward, emerging molecular imaging modalities using dyes conjugated with tumor-specific peptides may provide added sensitivity [8], [9], and [10]. It has to be noted that it is not possible to discriminate with certainty between primary landing sites and higher levels of drainage; however, the fact that fluorescent lymph nodes appeared 15–30 min after ICG injection and that direct drainage through lymphatic vessels could be seen in some cases suggests that these lymph nodes can be considered SLNs. Our study is limited by its small sample size. Nevertheless, it included a high number of lymph node metastases in the setting of extended PLND, allowing appropriate sensitivity analyses.

In conclusion, our lymphatic mapping study delineated a complex drainage pattern of the prostate that raises questions about whether targeted lymph node dissection can be implemented in PCa. Together with the low sensitivity for the detection of metastases, these results suggest that, for the time being, fluorescence SNL detection does not represent an alternative to a meticulously performed PLND in higher risk patients.

Author contributions: George N. Thalmann and Daniel P. Nguyen had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Nguyen, Thalmann.

Acquisition of data: Huber, Metzger, Schudel.

Analysis and interpretation of data: Nguyen, Huber.

Drafting of the manuscript: Nguyen.

Critical revision of the manuscript for important intellectual content: Huber, Metzger, Genitsch, Schudel, Thalmann.

Statistical analysis: Nguyen.

Obtaining funding: None.

Administrative, technical, or material support: None.

Supervision: Nguyen, Thalmann.

Other (specify): None.

Financial disclosures: George N. Thalmann and Daniel P. Nguyen certify that all conflicts of interest, including specific financial interests and relationships and affiliations relevant to the subject matter or materials discussed in the manuscript (eg, employment/ affiliation, grants or funding, consultancies, honoraria, stock ownership or options, expert testimony, royalties, or patents filed, received, or pending), are the following: None.

Funding/Support and role of the sponsor: The Fluobeam imaging device was provided at no charge by Fluoptics (Grenoble, France); however, the authors had complete control of the data and information submitted for publication.

The role of pelvic lymph node dissection (PLND) during radical prostatectomy (RP) remains a matter of continuous debate [1]. Sentinel lymph node (SLN) detection has been advanced as a potential alternative to PLND. In prostate cancer (PCa), the technique was first described using technetium Tc 99m bound to a colloid [2]; however, radioguided SLN detection has not come into widespread use. The use of the fluorescent dye indocyanine green (ICG) may open the door to broader acceptance of SLN techniques in PCa surgery [3]. Against this background, we provided a comprehensive description of lymphatic landing sites per anatomic region of the prostate using this SLN technique. We also evaluated the sensitivity of ICG-based fluorescence SLN detection to detect lymph node metastases in intermediate- and high-risk patients.

Detailed information on patient selection and detection technique is found in Supplement 1. From November 2012 through September 2015, 42 patients presenting with clinically localized intermediate- or high-risk PCa and scheduled for RP gave written informed consent to participate in our prospective study. ICG (Pulsion Medical Systems, Feldkirchen, Germany) was injected transrectally shortly prior to laparotomy. The first 12 patients enrolled received sextant injections (six injections into base, midportion, and apex of each prostatic lobe peripheral zone). The next 30 patients received injections into one of these sites: prostate base bilaterally, midportion bilaterally, apex bilaterally, left lobe (base, midportion, apex), or right lobe (base, midportion, apex). A near-infrared–sensitive probe (Fluobeam; Fluoptics, Grenoble, France) was used to collect fluorescence generated in the tissue under real-time image guidance. Independent of the findings of fluorescence SLN detection, an extended PLND was subsequently performed. An ex vivo fluorescence examination of all dissected lymph nodes was then carried out. Lymphatic landing sites per anatomic region of the prostate were depicted graphically. Diagnostic statistics assessed the value of ICG in detecting lymph node metastases.

Baseline characteristics of the patients are summarized in Supplementary Table 1. All 42 patients had one lymph node or more detected by fluorescence. The lymphatic mapping study showed the following results (Fig. 1; Supplementary Table 2): (1) The external and internal iliac regions encompassed the majority of SLNs; (2) the common iliac regions contained up to 22% of all SLNs; (3) a prostatic lobe can drain into the contralateral group of pelvic lymph nodes; (4) the fossa of Marcille also received significant drainage; and (5) practically all sites of the prostate can drain to different regions of the pelvis bilaterally.

Fig. 1

Percentages of sentinel nodes detected per drainage region with regard to anatomic sites of the prostate. The indocyanine green injection sites are depicted in the upper right corner. The orange zones represent the external iliac regions, the yellow zones represent the internal iliac regions, the green zones represent the common iliac regions, and the regions delineated by dashed lines represent the fossa of Marcille.

 

The drainage pattern did not show that distinct lymphatic pathways exist per prostatic anatomic region. Our results also underscored that lymphatics cross over to the opposite side and that the common iliac regions and the fossa of Marcille should not be overlooked during PLND, as combined, they may contain up to a third of all SLNs. The multitude of lymphatic landing sites as well as the individual variability of lymphatic drainage may represent obstacles to intraoperative SLN detection. In contrast to breast cancer, lymph node metastases in PCa do not follow a predefined pathway of metastatic spread, and there is no certainty that the histologic status of the SLN reflects the status of the entire pelvic node basin.

Among the 12 patients who received systematic ICG injections, a median of 15 SLNs per patient were removed (interquartile range [IQR]: 10–20). Five patients had a total of 29 lymph node metastases (Table 1). Of these, 16 were ICG positive, yielding a sensitivity for the detection of lymph node metastases of 55%. The negative predictive value was 95%, the specificity was 57%, and the predictive positive value was 8%. Repeating analyses with patients as the unit of analysis showed that four of five patients had ICG-positive lymph node metastases, yielding sensitivity of 80% for the correct staging of patients.

Hruby et al evaluated 38 patients who underwent laparoscopic RP [4]. The sensitivity of the SLN technique with ICG to detect metastases was 98% (42 nodes in 15 patients). Yuen et al reported on 66 patients who underwent RP and SNL detection using ICG [5]. Only nine metastases in six patients were detected, all of which were ICG positive. Our median number of nodes removed was 35 versus 18 in both previous reports. These findings suggest that our template leads to a notably more complete PLND, which would result in more precise sensitivity analyses. In addition, our median number of SLNs removed was 15 (IQR: 10–20), which is in the higher range of most previous studies using ICG alone or in combination with a radiocolloid [3], [4], [5], and [6]. Other authors may not have searched as thoroughly for SLNs in surgically less accessible regions such as the proximal part of the common iliac vessels, in the fossa of Marcille, and in the presacral regions. In the study by Yuen et al, only 2% and 1% of all SLNs were found in the common iliac and presacral regions, respectively [5].

The low sensitivity of ICG to detect metastases in the current study raises a concern about “skip” metastases. The presence of tumor cells in the lymph nodes may clog the feeding lymphatics and interfere with their ability to take up ICG, as has been postulated for radiocolloids [7]. Interestingly, we could sometimes visualize lymphatic drainage going around enlarged nodes. Going forward, emerging molecular imaging modalities using dyes conjugated with tumor-specific peptides may provide added sensitivity [8], [9], and [10]. It has to be noted that it is not possible to discriminate with certainty between primary landing sites and higher levels of drainage; however, the fact that fluorescent lymph nodes appeared 15–30 min after ICG injection and that direct drainage through lymphatic vessels could be seen in some cases suggests that these lymph nodes can be considered SLNs. Our study is limited by its small sample size. Nevertheless, it included a high number of lymph node metastases in the setting of extended PLND, allowing appropriate sensitivity analyses.

In conclusion, our lymphatic mapping study delineated a complex drainage pattern of the prostate that raises questions about whether targeted lymph node dissection can be implemented in PCa. Together with the low sensitivity for the detection of metastases, these results suggest that, for the time being, fluorescence SNL detection does not represent an alternative to a meticulously performed PLND in higher risk patients.

Author contributions: George N. Thalmann and Daniel P. Nguyen had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Nguyen, Thalmann.

Acquisition of data: Huber, Metzger, Schudel.

Analysis and interpretation of data: Nguyen, Huber.

Drafting of the manuscript: Nguyen.

Critical revision of the manuscript for important intellectual content: Huber, Metzger, Genitsch, Schudel, Thalmann.

Statistical analysis: Nguyen.

Obtaining funding: None.

Administrative, technical, or material support: None.

Supervision: Nguyen, Thalmann.

Other (specify): None.

Financial disclosures: George N. Thalmann and Daniel P. Nguyen certify that all conflicts of interest, including specific financial interests and relationships and affiliations relevant to the subject matter or materials discussed in the manuscript (eg, employment/ affiliation, grants or funding, consultancies, honoraria, stock ownership or options, expert testimony, royalties, or patents filed, received, or pending), are the following: None.

Funding/Support and role of the sponsor: The Fluobeam imaging device was provided at no charge by Fluoptics (Grenoble, France); however, the authors had complete control of the data and information submitted for publication.

The role of pelvic lymph node dissection (PLND) during radical prostatectomy (RP) remains a matter of continuous debate [1]. Sentinel lymph node (SLN) detection has been advanced as a potential alternative to PLND. In prostate cancer (PCa), the technique was first described using technetium Tc 99m bound to a colloid [2]; however, radioguided SLN detection has not come into widespread use. The use of the fluorescent dye indocyanine green (ICG) may open the door to broader acceptance of SLN techniques in PCa surgery [3]. Against this background, we provided a comprehensive description of lymphatic landing sites per anatomic region of the prostate using this SLN technique. We also evaluated the sensitivity of ICG-based fluorescence SLN detection to detect lymph node metastases in intermediate- and high-risk patients.

Detailed information on patient selection and detection technique is found in Supplement 1. From November 2012 through September 2015, 42 patients presenting with clinically localized intermediate- or high-risk PCa and scheduled for RP gave written informed consent to participate in our prospective study. ICG (Pulsion Medical Systems, Feldkirchen, Germany) was injected transrectally shortly prior to laparotomy. The first 12 patients enrolled received sextant injections (six injections into base, midportion, and apex of each prostatic lobe peripheral zone). The next 30 patients received injections into one of these sites: prostate base bilaterally, midportion bilaterally, apex bilaterally, left lobe (base, midportion, apex), or right lobe (base, midportion, apex). A near-infrared–sensitive probe (Fluobeam; Fluoptics, Grenoble, France) was used to collect fluorescence generated in the tissue under real-time image guidance. Independent of the findings of fluorescence SLN detection, an extended PLND was subsequently performed. An ex vivo fluorescence examination of all dissected lymph nodes was then carried out. Lymphatic landing sites per anatomic region of the prostate were depicted graphically. Diagnostic statistics assessed the value of ICG in detecting lymph node metastases.

Baseline characteristics of the patients are summarized in Supplementary Table 1. All 42 patients had one lymph node or more detected by fluorescence. The lymphatic mapping study showed the following results (Fig. 1; Supplementary Table 2): (1) The external and internal iliac regions encompassed the majority of SLNs; (2) the common iliac regions contained up to 22% of all SLNs; (3) a prostatic lobe can drain into the contralateral group of pelvic lymph nodes; (4) the fossa of Marcille also received significant drainage; and (5) practically all sites of the prostate can drain to different regions of the pelvis bilaterally.

Fig. 1

Percentages of sentinel nodes detected per drainage region with regard to anatomic sites of the prostate. The indocyanine green injection sites are depicted in the upper right corner. The orange zones represent the external iliac regions, the yellow zones represent the internal iliac regions, the green zones represent the common iliac regions, and the regions delineated by dashed lines represent the fossa of Marcille.

 

The drainage pattern did not show that distinct lymphatic pathways exist per prostatic anatomic region. Our results also underscored that lymphatics cross over to the opposite side and that the common iliac regions and the fossa of Marcille should not be overlooked during PLND, as combined, they may contain up to a third of all SLNs. The multitude of lymphatic landing sites as well as the individual variability of lymphatic drainage may represent obstacles to intraoperative SLN detection. In contrast to breast cancer, lymph node metastases in PCa do not follow a predefined pathway of metastatic spread, and there is no certainty that the histologic status of the SLN reflects the status of the entire pelvic node basin.

Among the 12 patients who received systematic ICG injections, a median of 15 SLNs per patient were removed (interquartile range [IQR]: 10–20). Five patients had a total of 29 lymph node metastases (Table 1). Of these, 16 were ICG positive, yielding a sensitivity for the detection of lymph node metastases of 55%. The negative predictive value was 95%, the specificity was 57%, and the predictive positive value was 8%. Repeating analyses with patients as the unit of analysis showed that four of five patients had ICG-positive lymph node metastases, yielding sensitivity of 80% for the correct staging of patients.

Hruby et al evaluated 38 patients who underwent laparoscopic RP [4]. The sensitivity of the SLN technique with ICG to detect metastases was 98% (42 nodes in 15 patients). Yuen et al reported on 66 patients who underwent RP and SNL detection using ICG [5]. Only nine metastases in six patients were detected, all of which were ICG positive. Our median number of nodes removed was 35 versus 18 in both previous reports. These findings suggest that our template leads to a notably more complete PLND, which would result in more precise sensitivity analyses. In addition, our median number of SLNs removed was 15 (IQR: 10–20), which is in the higher range of most previous studies using ICG alone or in combination with a radiocolloid [3], [4], [5], and [6]. Other authors may not have searched as thoroughly for SLNs in surgically less accessible regions such as the proximal part of the common iliac vessels, in the fossa of Marcille, and in the presacral regions. In the study by Yuen et al, only 2% and 1% of all SLNs were found in the common iliac and presacral regions, respectively [5].

The low sensitivity of ICG to detect metastases in the current study raises a concern about “skip” metastases. The presence of tumor cells in the lymph nodes may clog the feeding lymphatics and interfere with their ability to take up ICG, as has been postulated for radiocolloids [7]. Interestingly, we could sometimes visualize lymphatic drainage going around enlarged nodes. Going forward, emerging molecular imaging modalities using dyes conjugated with tumor-specific peptides may provide added sensitivity [8], [9], and [10]. It has to be noted that it is not possible to discriminate with certainty between primary landing sites and higher levels of drainage; however, the fact that fluorescent lymph nodes appeared 15–30 min after ICG injection and that direct drainage through lymphatic vessels could be seen in some cases suggests that these lymph nodes can be considered SLNs. Our study is limited by its small sample size. Nevertheless, it included a high number of lymph node metastases in the setting of extended PLND, allowing appropriate sensitivity analyses.

In conclusion, our lymphatic mapping study delineated a complex drainage pattern of the prostate that raises questions about whether targeted lymph node dissection can be implemented in PCa. Together with the low sensitivity for the detection of metastases, these results suggest that, for the time being, fluorescence SNL detection does not represent an alternative to a meticulously performed PLND in higher risk patients.

Author contributions: George N. Thalmann and Daniel P. Nguyen had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Nguyen, Thalmann.

Acquisition of data: Huber, Metzger, Schudel.

Analysis and interpretation of data: Nguyen, Huber.

Drafting of the manuscript: Nguyen.

Critical revision of the manuscript for important intellectual content: Huber, Metzger, Genitsch, Schudel, Thalmann.

Statistical analysis: Nguyen.

Obtaining funding: None.

Administrative, technical, or material support: None.

Supervision: Nguyen, Thalmann.

Other (specify): None.

Financial disclosures: George N. Thalmann and Daniel P. Nguyen certify that all conflicts of interest, including specific financial interests and relationships and affiliations relevant to the subject matter or materials discussed in the manuscript (eg, employment/ affiliation, grants or funding, consultancies, honoraria, stock ownership or options, expert testimony, royalties, or patents filed, received, or pending), are the following: None.

Funding/Support and role of the sponsor: The Fluobeam imaging device was provided at no charge by Fluoptics (Grenoble, France); however, the authors had complete control of the data and information submitted for publication.

The role of pelvic lymph node dissection (PLND) during radical prostatectomy (RP) remains a matter of continuous debate [1]. Sentinel lymph node (SLN) detection has been advanced as a potential alternative to PLND. In prostate cancer (PCa), the technique was first described using technetium Tc 99m bound to a colloid [2]; however, radioguided SLN detection has not come into widespread use. The use of the fluorescent dye indocyanine green (ICG) may open the door to broader acceptance of SLN techniques in PCa surgery [3]. Against this background, we provided a comprehensive description of lymphatic landing sites per anatomic region of the prostate using this SLN technique. We also evaluated the sensitivity of ICG-based fluorescence SLN detection to detect lymph node metastases in intermediate- and high-risk patients.

Detailed information on patient selection and detection technique is found in Supplement 1. From November 2012 through September 2015, 42 patients presenting with clinically localized intermediate- or high-risk PCa and scheduled for RP gave written informed consent to participate in our prospective study. ICG (Pulsion Medical Systems, Feldkirchen, Germany) was injected transrectally shortly prior to laparotomy. The first 12 patients enrolled received sextant injections (six injections into base, midportion, and apex of each prostatic lobe peripheral zone). The next 30 patients received injections into one of these sites: prostate base bilaterally, midportion bilaterally, apex bilaterally, left lobe (base, midportion, apex), or right lobe (base, midportion, apex). A near-infrared–sensitive probe (Fluobeam; Fluoptics, Grenoble, France) was used to collect fluorescence generated in the tissue under real-time image guidance. Independent of the findings of fluorescence SLN detection, an extended PLND was subsequently performed. An ex vivo fluorescence examination of all dissected lymph nodes was then carried out. Lymphatic landing sites per anatomic region of the prostate were depicted graphically. Diagnostic statistics assessed the value of ICG in detecting lymph node metastases.

Baseline characteristics of the patients are summarized in Supplementary Table 1. All 42 patients had one lymph node or more detected by fluorescence. The lymphatic mapping study showed the following results (Fig. 1; Supplementary Table 2): (1) The external and internal iliac regions encompassed the majority of SLNs; (2) the common iliac regions contained up to 22% of all SLNs; (3) a prostatic lobe can drain into the contralateral group of pelvic lymph nodes; (4) the fossa of Marcille also received significant drainage; and (5) practically all sites of the prostate can drain to different regions of the pelvis bilaterally.

Fig. 1

Percentages of sentinel nodes detected per drainage region with regard to anatomic sites of the prostate. The indocyanine green injection sites are depicted in the upper right corner. The orange zones represent the external iliac regions, the yellow zones represent the internal iliac regions, the green zones represent the common iliac regions, and the regions delineated by dashed lines represent the fossa of Marcille.

 

The drainage pattern did not show that distinct lymphatic pathways exist per prostatic anatomic region. Our results also underscored that lymphatics cross over to the opposite side and that the common iliac regions and the fossa of Marcille should not be overlooked during PLND, as combined, they may contain up to a third of all SLNs. The multitude of lymphatic landing sites as well as the individual variability of lymphatic drainage may represent obstacles to intraoperative SLN detection. In contrast to breast cancer, lymph node metastases in PCa do not follow a predefined pathway of metastatic spread, and there is no certainty that the histologic status of the SLN reflects the status of the entire pelvic node basin.

Among the 12 patients who received systematic ICG injections, a median of 15 SLNs per patient were removed (interquartile range [IQR]: 10–20). Five patients had a total of 29 lymph node metastases (Table 1). Of these, 16 were ICG positive, yielding a sensitivity for the detection of lymph node metastases of 55%. The negative predictive value was 95%, the specificity was 57%, and the predictive positive value was 8%. Repeating analyses with patients as the unit of analysis showed that four of five patients had ICG-positive lymph node metastases, yielding sensitivity of 80% for the correct staging of patients.

Hruby et al evaluated 38 patients who underwent laparoscopic RP [4]. The sensitivity of the SLN technique with ICG to detect metastases was 98% (42 nodes in 15 patients). Yuen et al reported on 66 patients who underwent RP and SNL detection using ICG [5]. Only nine metastases in six patients were detected, all of which were ICG positive. Our median number of nodes removed was 35 versus 18 in both previous reports. These findings suggest that our template leads to a notably more complete PLND, which would result in more precise sensitivity analyses. In addition, our median number of SLNs removed was 15 (IQR: 10–20), which is in the higher range of most previous studies using ICG alone or in combination with a radiocolloid [3], [4], [5], and [6]. Other authors may not have searched as thoroughly for SLNs in surgically less accessible regions such as the proximal part of the common iliac vessels, in the fossa of Marcille, and in the presacral regions. In the study by Yuen et al, only 2% and 1% of all SLNs were found in the common iliac and presacral regions, respectively [5].

The low sensitivity of ICG to detect metastases in the current study raises a concern about “skip” metastases. The presence of tumor cells in the lymph nodes may clog the feeding lymphatics and interfere with their ability to take up ICG, as has been postulated for radiocolloids [7]. Interestingly, we could sometimes visualize lymphatic drainage going around enlarged nodes. Going forward, emerging molecular imaging modalities using dyes conjugated with tumor-specific peptides may provide added sensitivity [8], [9], and [10]. It has to be noted that it is not possible to discriminate with certainty between primary landing sites and higher levels of drainage; however, the fact that fluorescent lymph nodes appeared 15–30 min after ICG injection and that direct drainage through lymphatic vessels could be seen in some cases suggests that these lymph nodes can be considered SLNs. Our study is limited by its small sample size. Nevertheless, it included a high number of lymph node metastases in the setting of extended PLND, allowing appropriate sensitivity analyses.

In conclusion, our lymphatic mapping study delineated a complex drainage pattern of the prostate that raises questions about whether targeted lymph node dissection can be implemented in PCa. Together with the low sensitivity for the detection of metastases, these results suggest that, for the time being, fluorescence SNL detection does not represent an alternative to a meticulously performed PLND in higher risk patients.

Author contributions: George N. Thalmann and Daniel P. Nguyen had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Nguyen, Thalmann.

Acquisition of data: Huber, Metzger, Schudel.

Analysis and interpretation of data: Nguyen, Huber.

Drafting of the manuscript: Nguyen.

Critical revision of the manuscript for important intellectual content: Huber, Metzger, Genitsch, Schudel, Thalmann.

Statistical analysis: Nguyen.

Obtaining funding: None.

Administrative, technical, or material support: None.

Supervision: Nguyen, Thalmann.

Other (specify): None.

Financial disclosures: George N. Thalmann and Daniel P. Nguyen certify that all conflicts of interest, including specific financial interests and relationships and affiliations relevant to the subject matter or materials discussed in the manuscript (eg, employment/ affiliation, grants or funding, consultancies, honoraria, stock ownership or options, expert testimony, royalties, or patents filed, received, or pending), are the following: None.

Funding/Support and role of the sponsor: The Fluobeam imaging device was provided at no charge by Fluoptics (Grenoble, France); however, the authors had complete control of the data and information submitted for publication.

The role of pelvic lymph node dissection (PLND) during radical prostatectomy (RP) remains a matter of continuous debate [1]. Sentinel lymph node (SLN) detection has been advanced as a potential alternative to PLND. In prostate cancer (PCa), the technique was first described using technetium Tc 99m bound to a colloid [2]; however, radioguided SLN detection has not come into widespread use. The use of the fluorescent dye indocyanine green (ICG) may open the door to broader acceptance of SLN techniques in PCa surgery [3]. Against this background, we provided a comprehensive description of lymphatic landing sites per anatomic region of the prostate using this SLN technique. We also evaluated the sensitivity of ICG-based fluorescence SLN detection to detect lymph node metastases in intermediate- and high-risk patients.

Detailed information on patient selection and detection technique is found in Supplement 1. From November 2012 through September 2015, 42 patients presenting with clinically localized intermediate- or high-risk PCa and scheduled for RP gave written informed consent to participate in our prospective study. ICG (Pulsion Medical Systems, Feldkirchen, Germany) was injected transrectally shortly prior to laparotomy. The first 12 patients enrolled received sextant injections (six injections into base, midportion, and apex of each prostatic lobe peripheral zone). The next 30 patients received injections into one of these sites: prostate base bilaterally, midportion bilaterally, apex bilaterally, left lobe (base, midportion, apex), or right lobe (base, midportion, apex). A near-infrared–sensitive probe (Fluobeam; Fluoptics, Grenoble, France) was used to collect fluorescence generated in the tissue under real-time image guidance. Independent of the findings of fluorescence SLN detection, an extended PLND was subsequently performed. An ex vivo fluorescence examination of all dissected lymph nodes was then carried out. Lymphatic landing sites per anatomic region of the prostate were depicted graphically. Diagnostic statistics assessed the value of ICG in detecting lymph node metastases.

Baseline characteristics of the patients are summarized in Supplementary Table 1. All 42 patients had one lymph node or more detected by fluorescence. The lymphatic mapping study showed the following results (Fig. 1; Supplementary Table 2): (1) The external and internal iliac regions encompassed the majority of SLNs; (2) the common iliac regions contained up to 22% of all SLNs; (3) a prostatic lobe can drain into the contralateral group of pelvic lymph nodes; (4) the fossa of Marcille also received significant drainage; and (5) practically all sites of the prostate can drain to different regions of the pelvis bilaterally.

Fig. 1

Percentages of sentinel nodes detected per drainage region with regard to anatomic sites of the prostate. The indocyanine green injection sites are depicted in the upper right corner. The orange zones represent the external iliac regions, the yellow zones represent the internal iliac regions, the green zones represent the common iliac regions, and the regions delineated by dashed lines represent the fossa of Marcille.

 

The drainage pattern did not show that distinct lymphatic pathways exist per prostatic anatomic region. Our results also underscored that lymphatics cross over to the opposite side and that the common iliac regions and the fossa of Marcille should not be overlooked during PLND, as combined, they may contain up to a third of all SLNs. The multitude of lymphatic landing sites as well as the individual variability of lymphatic drainage may represent obstacles to intraoperative SLN detection. In contrast to breast cancer, lymph node metastases in PCa do not follow a predefined pathway of metastatic spread, and there is no certainty that the histologic status of the SLN reflects the status of the entire pelvic node basin.

Among the 12 patients who received systematic ICG injections, a median of 15 SLNs per patient were removed (interquartile range [IQR]: 10–20). Five patients had a total of 29 lymph node metastases (Table 1). Of these, 16 were ICG positive, yielding a sensitivity for the detection of lymph node metastases of 55%. The negative predictive value was 95%, the specificity was 57%, and the predictive positive value was 8%. Repeating analyses with patients as the unit of analysis showed that four of five patients had ICG-positive lymph node metastases, yielding sensitivity of 80% for the correct staging of patients.

Hruby et al evaluated 38 patients who underwent laparoscopic RP [4]. The sensitivity of the SLN technique with ICG to detect metastases was 98% (42 nodes in 15 patients). Yuen et al reported on 66 patients who underwent RP and SNL detection using ICG [5]. Only nine metastases in six patients were detected, all of which were ICG positive. Our median number of nodes removed was 35 versus 18 in both previous reports. These findings suggest that our template leads to a notably more complete PLND, which would result in more precise sensitivity analyses. In addition, our median number of SLNs removed was 15 (IQR: 10–20), which is in the higher range of most previous studies using ICG alone or in combination with a radiocolloid [3], [4], [5], and [6]. Other authors may not have searched as thoroughly for SLNs in surgically less accessible regions such as the proximal part of the common iliac vessels, in the fossa of Marcille, and in the presacral regions. In the study by Yuen et al, only 2% and 1% of all SLNs were found in the common iliac and presacral regions, respectively [5].

The low sensitivity of ICG to detect metastases in the current study raises a concern about “skip” metastases. The presence of tumor cells in the lymph nodes may clog the feeding lymphatics and interfere with their ability to take up ICG, as has been postulated for radiocolloids [7]. Interestingly, we could sometimes visualize lymphatic drainage going around enlarged nodes. Going forward, emerging molecular imaging modalities using dyes conjugated with tumor-specific peptides may provide added sensitivity [8], [9], and [10]. It has to be noted that it is not possible to discriminate with certainty between primary landing sites and higher levels of drainage; however, the fact that fluorescent lymph nodes appeared 15–30 min after ICG injection and that direct drainage through lymphatic vessels could be seen in some cases suggests that these lymph nodes can be considered SLNs. Our study is limited by its small sample size. Nevertheless, it included a high number of lymph node metastases in the setting of extended PLND, allowing appropriate sensitivity analyses.

In conclusion, our lymphatic mapping study delineated a complex drainage pattern of the prostate that raises questions about whether targeted lymph node dissection can be implemented in PCa. Together with the low sensitivity for the detection of metastases, these results suggest that, for the time being, fluorescence SNL detection does not represent an alternative to a meticulously performed PLND in higher risk patients.

Author contributions: George N. Thalmann and Daniel P. Nguyen had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Nguyen, Thalmann.

Acquisition of data: Huber, Metzger, Schudel.

Analysis and interpretation of data: Nguyen, Huber.

Drafting of the manuscript: Nguyen.

Critical revision of the manuscript for important intellectual content: Huber, Metzger, Genitsch, Schudel, Thalmann.

Statistical analysis: Nguyen.

Obtaining funding: None.

Administrative, technical, or material support: None.

Supervision: Nguyen, Thalmann.

Other (specify): None.

Financial disclosures: George N. Thalmann and Daniel P. Nguyen certify that all conflicts of interest, including specific financial interests and relationships and affiliations relevant to the subject matter or materials discussed in the manuscript (eg, employment/ affiliation, grants or funding, consultancies, honoraria, stock ownership or options, expert testimony, royalties, or patents filed, received, or pending), are the following: None.

Funding/Support and role of the sponsor: The Fluobeam imaging device was provided at no charge by Fluoptics (Grenoble, France); however, the authors had complete control of the data and information submitted for publication.

The role of pelvic lymph node dissection (PLND) during radical prostatectomy (RP) remains a matter of continuous debate [1]. Sentinel lymph node (SLN) detection has been advanced as a potential alternative to PLND. In prostate cancer (PCa), the technique was first described using technetium Tc 99m bound to a colloid [2]; however, radioguided SLN detection has not come into widespread use. The use of the fluorescent dye indocyanine green (ICG) may open the door to broader acceptance of SLN techniques in PCa surgery [3]. Against this background, we provided a comprehensive description of lymphatic landing sites per anatomic region of the prostate using this SLN technique. We also evaluated the sensitivity of ICG-based fluorescence SLN detection to detect lymph node metastases in intermediate- and high-risk patients.

Detailed information on patient selection and detection technique is found in Supplement 1. From November 2012 through September 2015, 42 patients presenting with clinically localized intermediate- or high-risk PCa and scheduled for RP gave written informed consent to participate in our prospective study. ICG (Pulsion Medical Systems, Feldkirchen, Germany) was injected transrectally shortly prior to laparotomy. The first 12 patients enrolled received sextant injections (six injections into base, midportion, and apex of each prostatic lobe peripheral zone). The next 30 patients received injections into one of these sites: prostate base bilaterally, midportion bilaterally, apex bilaterally, left lobe (base, midportion, apex), or right lobe (base, midportion, apex). A near-infrared–sensitive probe (Fluobeam; Fluoptics, Grenoble, France) was used to collect fluorescence generated in the tissue under real-time image guidance. Independent of the findings of fluorescence SLN detection, an extended PLND was subsequently performed. An ex vivo fluorescence examination of all dissected lymph nodes was then carried out. Lymphatic landing sites per anatomic region of the prostate were depicted graphically. Diagnostic statistics assessed the value of ICG in detecting lymph node metastases.

Baseline characteristics of the patients are summarized in Supplementary Table 1. All 42 patients had one lymph node or more detected by fluorescence. The lymphatic mapping study showed the following results (Fig. 1; Supplementary Table 2): (1) The external and internal iliac regions encompassed the majority of SLNs; (2) the common iliac regions contained up to 22% of all SLNs; (3) a prostatic lobe can drain into the contralateral group of pelvic lymph nodes; (4) the fossa of Marcille also received significant drainage; and (5) practically all sites of the prostate can drain to different regions of the pelvis bilaterally.

Fig. 1

Percentages of sentinel nodes detected per drainage region with regard to anatomic sites of the prostate. The indocyanine green injection sites are depicted in the upper right corner. The orange zones represent the external iliac regions, the yellow zones represent the internal iliac regions, the green zones represent the common iliac regions, and the regions delineated by dashed lines represent the fossa of Marcille.

 

The drainage pattern did not show that distinct lymphatic pathways exist per prostatic anatomic region. Our results also underscored that lymphatics cross over to the opposite side and that the common iliac regions and the fossa of Marcille should not be overlooked during PLND, as combined, they may contain up to a third of all SLNs. The multitude of lymphatic landing sites as well as the individual variability of lymphatic drainage may represent obstacles to intraoperative SLN detection. In contrast to breast cancer, lymph node metastases in PCa do not follow a predefined pathway of metastatic spread, and there is no certainty that the histologic status of the SLN reflects the status of the entire pelvic node basin.

Among the 12 patients who received systematic ICG injections, a median of 15 SLNs per patient were removed (interquartile range [IQR]: 10–20). Five patients had a total of 29 lymph node metastases (Table 1). Of these, 16 were ICG positive, yielding a sensitivity for the detection of lymph node metastases of 55%. The negative predictive value was 95%, the specificity was 57%, and the predictive positive value was 8%. Repeating analyses with patients as the unit of analysis showed that four of five patients had ICG-positive lymph node metastases, yielding sensitivity of 80% for the correct staging of patients.

Hruby et al evaluated 38 patients who underwent laparoscopic RP [4]. The sensitivity of the SLN technique with ICG to detect metastases was 98% (42 nodes in 15 patients). Yuen et al reported on 66 patients who underwent RP and SNL detection using ICG [5]. Only nine metastases in six patients were detected, all of which were ICG positive. Our median number of nodes removed was 35 versus 18 in both previous reports. These findings suggest that our template leads to a notably more complete PLND, which would result in more precise sensitivity analyses. In addition, our median number of SLNs removed was 15 (IQR: 10–20), which is in the higher range of most previous studies using ICG alone or in combination with a radiocolloid [3], [4], [5], and [6]. Other authors may not have searched as thoroughly for SLNs in surgically less accessible regions such as the proximal part of the common iliac vessels, in the fossa of Marcille, and in the presacral regions. In the study by Yuen et al, only 2% and 1% of all SLNs were found in the common iliac and presacral regions, respectively [5].

The low sensitivity of ICG to detect metastases in the current study raises a concern about “skip” metastases. The presence of tumor cells in the lymph nodes may clog the feeding lymphatics and interfere with their ability to take up ICG, as has been postulated for radiocolloids [7]. Interestingly, we could sometimes visualize lymphatic drainage going around enlarged nodes. Going forward, emerging molecular imaging modalities using dyes conjugated with tumor-specific peptides may provide added sensitivity [8], [9], and [10]. It has to be noted that it is not possible to discriminate with certainty between primary landing sites and higher levels of drainage; however, the fact that fluorescent lymph nodes appeared 15–30 min after ICG injection and that direct drainage through lymphatic vessels could be seen in some cases suggests that these lymph nodes can be considered SLNs. Our study is limited by its small sample size. Nevertheless, it included a high number of lymph node metastases in the setting of extended PLND, allowing appropriate sensitivity analyses.

In conclusion, our lymphatic mapping study delineated a complex drainage pattern of the prostate that raises questions about whether targeted lymph node dissection can be implemented in PCa. Together with the low sensitivity for the detection of metastases, these results suggest that, for the time being, fluorescence SNL detection does not represent an alternative to a meticulously performed PLND in higher risk patients.

Author contributions: George N. Thalmann and Daniel P. Nguyen had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Nguyen, Thalmann.

Acquisition of data: Huber, Metzger, Schudel.

Analysis and interpretation of data: Nguyen, Huber.

Drafting of the manuscript: Nguyen.

Critical revision of the manuscript for important intellectual content: Huber, Metzger, Genitsch, Schudel, Thalmann.

Statistical analysis: Nguyen.

Obtaining funding: None.

Administrative, technical, or material support: None.

Supervision: Nguyen, Thalmann.

Other (specify): None.

Financial disclosures: George N. Thalmann and Daniel P. Nguyen certify that all conflicts of interest, including specific financial interests and relationships and affiliations relevant to the subject matter or materials discussed in the manuscript (eg, employment/ affiliation, grants or funding, consultancies, honoraria, stock ownership or options, expert testimony, royalties, or patents filed, received, or pending), are the following: None.

Funding/Support and role of the sponsor: The Fluobeam imaging device was provided at no charge by Fluoptics (Grenoble, France); however, the authors had complete control of the data and information submitted for publication.

The role of pelvic lymph node dissection (PLND) during radical prostatectomy (RP) remains a matter of continuous debate [1]. Sentinel lymph node (SLN) detection has been advanced as a potential alternative to PLND. In prostate cancer (PCa), the technique was first described using technetium Tc 99m bound to a colloid [2]; however, radioguided SLN detection has not come into widespread use. The use of the fluorescent dye indocyanine green (ICG) may open the door to broader acceptance of SLN techniques in PCa surgery [3]. Against this background, we provided a comprehensive description of lymphatic landing sites per anatomic region of the prostate using this SLN technique. We also evaluated the sensitivity of ICG-based fluorescence SLN detection to detect lymph node metastases in intermediate- and high-risk patients.

Detailed information on patient selection and detection technique is found in Supplement 1. From November 2012 through September 2015, 42 patients presenting with clinically localized intermediate- or high-risk PCa and scheduled for RP gave written informed consent to participate in our prospective study. ICG (Pulsion Medical Systems, Feldkirchen, Germany) was injected transrectally shortly prior to laparotomy. The first 12 patients enrolled received sextant injections (six injections into base, midportion, and apex of each prostatic lobe peripheral zone). The next 30 patients received injections into one of these sites: prostate base bilaterally, midportion bilaterally, apex bilaterally, left lobe (base, midportion, apex), or right lobe (base, midportion, apex). A near-infrared–sensitive probe (Fluobeam; Fluoptics, Grenoble, France) was used to collect fluorescence generated in the tissue under real-time image guidance. Independent of the findings of fluorescence SLN detection, an extended PLND was subsequently performed. An ex vivo fluorescence examination of all dissected lymph nodes was then carried out. Lymphatic landing sites per anatomic region of the prostate were depicted graphically. Diagnostic statistics assessed the value of ICG in detecting lymph node metastases.

Baseline characteristics of the patients are summarized in Supplementary Table 1. All 42 patients had one lymph node or more detected by fluorescence. The lymphatic mapping study showed the following results (Fig. 1; Supplementary Table 2): (1) The external and internal iliac regions encompassed the majority of SLNs; (2) the common iliac regions contained up to 22% of all SLNs; (3) a prostatic lobe can drain into the contralateral group of pelvic lymph nodes; (4) the fossa of Marcille also received significant drainage; and (5) practically all sites of the prostate can drain to different regions of the pelvis bilaterally.

Fig. 1

Percentages of sentinel nodes detected per drainage region with regard to anatomic sites of the prostate. The indocyanine green injection sites are depicted in the upper right corner. The orange zones represent the external iliac regions, the yellow zones represent the internal iliac regions, the green zones represent the common iliac regions, and the regions delineated by dashed lines represent the fossa of Marcille.

 

The drainage pattern did not show that distinct lymphatic pathways exist per prostatic anatomic region. Our results also underscored that lymphatics cross over to the opposite side and that the common iliac regions and the fossa of Marcille should not be overlooked during PLND, as combined, they may contain up to a third of all SLNs. The multitude of lymphatic landing sites as well as the individual variability of lymphatic drainage may represent obstacles to intraoperative SLN detection. In contrast to breast cancer, lymph node metastases in PCa do not follow a predefined pathway of metastatic spread, and there is no certainty that the histologic status of the SLN reflects the status of the entire pelvic node basin.

Among the 12 patients who received systematic ICG injections, a median of 15 SLNs per patient were removed (interquartile range [IQR]: 10–20). Five patients had a total of 29 lymph node metastases (Table 1). Of these, 16 were ICG positive, yielding a sensitivity for the detection of lymph node metastases of 55%. The negative predictive value was 95%, the specificity was 57%, and the predictive positive value was 8%. Repeating analyses with patients as the unit of analysis showed that four of five patients had ICG-positive lymph node metastases, yielding sensitivity of 80% for the correct staging of patients.

Hruby et al evaluated 38 patients who underwent laparoscopic RP [4]. The sensitivity of the SLN technique with ICG to detect metastases was 98% (42 nodes in 15 patients). Yuen et al reported on 66 patients who underwent RP and SNL detection using ICG [5]. Only nine metastases in six patients were detected, all of which were ICG positive. Our median number of nodes removed was 35 versus 18 in both previous reports. These findings suggest that our template leads to a notably more complete PLND, which would result in more precise sensitivity analyses. In addition, our median number of SLNs removed was 15 (IQR: 10–20), which is in the higher range of most previous studies using ICG alone or in combination with a radiocolloid [3], [4], [5], and [6]. Other authors may not have searched as thoroughly for SLNs in surgically less accessible regions such as the proximal part of the common iliac vessels, in the fossa of Marcille, and in the presacral regions. In the study by Yuen et al, only 2% and 1% of all SLNs were found in the common iliac and presacral regions, respectively [5].

The low sensitivity of ICG to detect metastases in the current study raises a concern about “skip” metastases. The presence of tumor cells in the lymph nodes may clog the feeding lymphatics and interfere with their ability to take up ICG, as has been postulated for radiocolloids [7]. Interestingly, we could sometimes visualize lymphatic drainage going around enlarged nodes. Going forward, emerging molecular imaging modalities using dyes conjugated with tumor-specific peptides may provide added sensitivity [8], [9], and [10]. It has to be noted that it is not possible to discriminate with certainty between primary landing sites and higher levels of drainage; however, the fact that fluorescent lymph nodes appeared 15–30 min after ICG injection and that direct drainage through lymphatic vessels could be seen in some cases suggests that these lymph nodes can be considered SLNs. Our study is limited by its small sample size. Nevertheless, it included a high number of lymph node metastases in the setting of extended PLND, allowing appropriate sensitivity analyses.

In conclusion, our lymphatic mapping study delineated a complex drainage pattern of the prostate that raises questions about whether targeted lymph node dissection can be implemented in PCa. Together with the low sensitivity for the detection of metastases, these results suggest that, for the time being, fluorescence SNL detection does not represent an alternative to a meticulously performed PLND in higher risk patients.

Author contributions: George N. Thalmann and Daniel P. Nguyen had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Nguyen, Thalmann.

Acquisition of data: Huber, Metzger, Schudel.

Analysis and interpretation of data: Nguyen, Huber.

Drafting of the manuscript: Nguyen.

Critical revision of the manuscript for important intellectual content: Huber, Metzger, Genitsch, Schudel, Thalmann.

Statistical analysis: Nguyen.

Obtaining funding: None.

Administrative, technical, or material support: None.

Supervision: Nguyen, Thalmann.

Other (specify): None.

Financial disclosures: George N. Thalmann and Daniel P. Nguyen certify that all conflicts of interest, including specific financial interests and relationships and affiliations relevant to the subject matter or materials discussed in the manuscript (eg, employment/ affiliation, grants or funding, consultancies, honoraria, stock ownership or options, expert testimony, royalties, or patents filed, received, or pending), are the following: None.

Funding/Support and role of the sponsor: The Fluobeam imaging device was provided at no charge by Fluoptics (Grenoble, France); however, the authors had complete control of the data and information submitted for publication.

The role of pelvic lymph node dissection (PLND) during radical prostatectomy (RP) remains a matter of continuous debate [1]. Sentinel lymph node (SLN) detection has been advanced as a potential alternative to PLND. In prostate cancer (PCa), the technique was first described using technetium Tc 99m bound to a colloid [2]; however, radioguided SLN detection has not come into widespread use. The use of the fluorescent dye indocyanine green (ICG) may open the door to broader acceptance of SLN techniques in PCa surgery [3]. Against this background, we provided a comprehensive description of lymphatic landing sites per anatomic region of the prostate using this SLN technique. We also evaluated the sensitivity of ICG-based fluorescence SLN detection to detect lymph node metastases in intermediate- and high-risk patients.

Detailed information on patient selection and detection technique is found in Supplement 1. From November 2012 through September 2015, 42 patients presenting with clinically localized intermediate- or high-risk PCa and scheduled for RP gave written informed consent to participate in our prospective study. ICG (Pulsion Medical Systems, Feldkirchen, Germany) was injected transrectally shortly prior to laparotomy. The first 12 patients enrolled received sextant injections (six injections into base, midportion, and apex of each prostatic lobe peripheral zone). The next 30 patients received injections into one of these sites: prostate base bilaterally, midportion bilaterally, apex bilaterally, left lobe (base, midportion, apex), or right lobe (base, midportion, apex). A near-infrared–sensitive probe (Fluobeam; Fluoptics, Grenoble, France) was used to collect fluorescence generated in the tissue under real-time image guidance. Independent of the findings of fluorescence SLN detection, an extended PLND was subsequently performed. An ex vivo fluorescence examination of all dissected lymph nodes was then carried out. Lymphatic landing sites per anatomic region of the prostate were depicted graphically. Diagnostic statistics assessed the value of ICG in detecting lymph node metastases.

Baseline characteristics of the patients are summarized in Supplementary Table 1. All 42 patients had one lymph node or more detected by fluorescence. The lymphatic mapping study showed the following results (Fig. 1; Supplementary Table 2): (1) The external and internal iliac regions encompassed the majority of SLNs; (2) the common iliac regions contained up to 22% of all SLNs; (3) a prostatic lobe can drain into the contralateral group of pelvic lymph nodes; (4) the fossa of Marcille also received significant drainage; and (5) practically all sites of the prostate can drain to different regions of the pelvis bilaterally.

Fig. 1

Percentages of sentinel nodes detected per drainage region with regard to anatomic sites of the prostate. The indocyanine green injection sites are depicted in the upper right corner. The orange zones represent the external iliac regions, the yellow zones represent the internal iliac regions, the green zones represent the common iliac regions, and the regions delineated by dashed lines represent the fossa of Marcille.

 

The drainage pattern did not show that distinct lymphatic pathways exist per prostatic anatomic region. Our results also underscored that lymphatics cross over to the opposite side and that the common iliac regions and the fossa of Marcille should not be overlooked during PLND, as combined, they may contain up to a third of all SLNs. The multitude of lymphatic landing sites as well as the individual variability of lymphatic drainage may represent obstacles to intraoperative SLN detection. In contrast to breast cancer, lymph node metastases in PCa do not follow a predefined pathway of metastatic spread, and there is no certainty that the histologic status of the SLN reflects the status of the entire pelvic node basin.

Among the 12 patients who received systematic ICG injections, a median of 15 SLNs per patient were removed (interquartile range [IQR]: 10–20). Five patients had a total of 29 lymph node metastases (Table 1). Of these, 16 were ICG positive, yielding a sensitivity for the detection of lymph node metastases of 55%. The negative predictive value was 95%, the specificity was 57%, and the predictive positive value was 8%. Repeating analyses with patients as the unit of analysis showed that four of five patients had ICG-positive lymph node metastases, yielding sensitivity of 80% for the correct staging of patients.

Hruby et al evaluated 38 patients who underwent laparoscopic RP [4]. The sensitivity of the SLN technique with ICG to detect metastases was 98% (42 nodes in 15 patients). Yuen et al reported on 66 patients who underwent RP and SNL detection using ICG [5]. Only nine metastases in six patients were detected, all of which were ICG positive. Our median number of nodes removed was 35 versus 18 in both previous reports. These findings suggest that our template leads to a notably more complete PLND, which would result in more precise sensitivity analyses. In addition, our median number of SLNs removed was 15 (IQR: 10–20), which is in the higher range of most previous studies using ICG alone or in combination with a radiocolloid [3], [4], [5], and [6]. Other authors may not have searched as thoroughly for SLNs in surgically less accessible regions such as the proximal part of the common iliac vessels, in the fossa of Marcille, and in the presacral regions. In the study by Yuen et al, only 2% and 1% of all SLNs were found in the common iliac and presacral regions, respectively [5].

The low sensitivity of ICG to detect metastases in the current study raises a concern about “skip” metastases. The presence of tumor cells in the lymph nodes may clog the feeding lymphatics and interfere with their ability to take up ICG, as has been postulated for radiocolloids [7]. Interestingly, we could sometimes visualize lymphatic drainage going around enlarged nodes. Going forward, emerging molecular imaging modalities using dyes conjugated with tumor-specific peptides may provide added sensitivity [8], [9], and [10]. It has to be noted that it is not possible to discriminate with certainty between primary landing sites and higher levels of drainage; however, the fact that fluorescent lymph nodes appeared 15–30 min after ICG injection and that direct drainage through lymphatic vessels could be seen in some cases suggests that these lymph nodes can be considered SLNs. Our study is limited by its small sample size. Nevertheless, it included a high number of lymph node metastases in the setting of extended PLND, allowing appropriate sensitivity analyses.

In conclusion, our lymphatic mapping study delineated a complex drainage pattern of the prostate that raises questions about whether targeted lymph node dissection can be implemented in PCa. Together with the low sensitivity for the detection of metastases, these results suggest that, for the time being, fluorescence SNL detection does not represent an alternative to a meticulously performed PLND in higher risk patients.

Author contributions: George N. Thalmann and Daniel P. Nguyen had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Nguyen, Thalmann.

Acquisition of data: Huber, Metzger, Schudel.

Analysis and interpretation of data: Nguyen, Huber.

Drafting of the manuscript: Nguyen.

Critical revision of the manuscript for important intellectual content: Huber, Metzger, Genitsch, Schudel, Thalmann.

Statistical analysis: Nguyen.

Obtaining funding: None.

Administrative, technical, or material support: None.

Supervision: Nguyen, Thalmann.

Other (specify): None.

Financial disclosures: George N. Thalmann and Daniel P. Nguyen certify that all conflicts of interest, including specific financial interests and relationships and affiliations relevant to the subject matter or materials discussed in the manuscript (eg, employment/ affiliation, grants or funding, consultancies, honoraria, stock ownership or options, expert testimony, royalties, or patents filed, received, or pending), are the following: None.

Funding/Support and role of the sponsor: The Fluobeam imaging device was provided at no charge by Fluoptics (Grenoble, France); however, the authors had complete control of the data and information submitted for publication.

The role of pelvic lymph node dissection (PLND) during radical prostatectomy (RP) remains a matter of continuous debate [1]. Sentinel lymph node (SLN) detection has been advanced as a potential alternative to PLND. In prostate cancer (PCa), the technique was first described using technetium Tc 99m bound to a colloid [2]; however, radioguided SLN detection has not come into widespread use. The use of the fluorescent dye indocyanine green (ICG) may open the door to broader acceptance of SLN techniques in PCa surgery [3]. Against this background, we provided a comprehensive description of lymphatic landing sites per anatomic region of the prostate using this SLN technique. We also evaluated the sensitivity of ICG-based fluorescence SLN detection to detect lymph node metastases in intermediate- and high-risk patients.

Detailed information on patient selection and detection technique is found in Supplement 1. From November 2012 through September 2015, 42 patients presenting with clinically localized intermediate- or high-risk PCa and scheduled for RP gave written informed consent to participate in our prospective study. ICG (Pulsion Medical Systems, Feldkirchen, Germany) was injected transrectally shortly prior to laparotomy. The first 12 patients enrolled received sextant injections (six injections into base, midportion, and apex of each prostatic lobe peripheral zone). The next 30 patients received injections into one of these sites: prostate base bilaterally, midportion bilaterally, apex bilaterally, left lobe (base, midportion, apex), or right lobe (base, midportion, apex). A near-infrared–sensitive probe (Fluobeam; Fluoptics, Grenoble, France) was used to collect fluorescence generated in the tissue under real-time image guidance. Independent of the findings of fluorescence SLN detection, an extended PLND was subsequently performed. An ex vivo fluorescence examination of all dissected lymph nodes was then carried out. Lymphatic landing sites per anatomic region of the prostate were depicted graphically. Diagnostic statistics assessed the value of ICG in detecting lymph node metastases.

Baseline characteristics of the patients are summarized in Supplementary Table 1. All 42 patients had one lymph node or more detected by fluorescence. The lymphatic mapping study showed the following results (Fig. 1; Supplementary Table 2): (1) The external and internal iliac regions encompassed the majority of SLNs; (2) the common iliac regions contained up to 22% of all SLNs; (3) a prostatic lobe can drain into the contralateral group of pelvic lymph nodes; (4) the fossa of Marcille also received significant drainage; and (5) practically all sites of the prostate can drain to different regions of the pelvis bilaterally.

Fig. 1

Percentages of sentinel nodes detected per drainage region with regard to anatomic sites of the prostate. The indocyanine green injection sites are depicted in the upper right corner. The orange zones represent the external iliac regions, the yellow zones represent the internal iliac regions, the green zones represent the common iliac regions, and the regions delineated by dashed lines represent the fossa of Marcille.

 

The drainage pattern did not show that distinct lymphatic pathways exist per prostatic anatomic region. Our results also underscored that lymphatics cross over to the opposite side and that the common iliac regions and the fossa of Marcille should not be overlooked during PLND, as combined, they may contain up to a third of all SLNs. The multitude of lymphatic landing sites as well as the individual variability of lymphatic drainage may represent obstacles to intraoperative SLN detection. In contrast to breast cancer, lymph node metastases in PCa do not follow a predefined pathway of metastatic spread, and there is no certainty that the histologic status of the SLN reflects the status of the entire pelvic node basin.

Among the 12 patients who received systematic ICG injections, a median of 15 SLNs per patient were removed (interquartile range [IQR]: 10–20). Five patients had a total of 29 lymph node metastases (Table 1). Of these, 16 were ICG positive, yielding a sensitivity for the detection of lymph node metastases of 55%. The negative predictive value was 95%, the specificity was 57%, and the predictive positive value was 8%. Repeating analyses with patients as the unit of analysis showed that four of five patients had ICG-positive lymph node metastases, yielding sensitivity of 80% for the correct staging of patients.

Hruby et al evaluated 38 patients who underwent laparoscopic RP [4]. The sensitivity of the SLN technique with ICG to detect metastases was 98% (42 nodes in 15 patients). Yuen et al reported on 66 patients who underwent RP and SNL detection using ICG [5]. Only nine metastases in six patients were detected, all of which were ICG positive. Our median number of nodes removed was 35 versus 18 in both previous reports. These findings suggest that our template leads to a notably more complete PLND, which would result in more precise sensitivity analyses. In addition, our median number of SLNs removed was 15 (IQR: 10–20), which is in the higher range of most previous studies using ICG alone or in combination with a radiocolloid [3], [4], [5], and [6]. Other authors may not have searched as thoroughly for SLNs in surgically less accessible regions such as the proximal part of the common iliac vessels, in the fossa of Marcille, and in the presacral regions. In the study by Yuen et al, only 2% and 1% of all SLNs were found in the common iliac and presacral regions, respectively [5].

The low sensitivity of ICG to detect metastases in the current study raises a concern about “skip” metastases. The presence of tumor cells in the lymph nodes may clog the feeding lymphatics and interfere with their ability to take up ICG, as has been postulated for radiocolloids [7]. Interestingly, we could sometimes visualize lymphatic drainage going around enlarged nodes. Going forward, emerging molecular imaging modalities using dyes conjugated with tumor-specific peptides may provide added sensitivity [8], [9], and [10]. It has to be noted that it is not possible to discriminate with certainty between primary landing sites and higher levels of drainage; however, the fact that fluorescent lymph nodes appeared 15–30 min after ICG injection and that direct drainage through lymphatic vessels could be seen in some cases suggests that these lymph nodes can be considered SLNs. Our study is limited by its small sample size. Nevertheless, it included a high number of lymph node metastases in the setting of extended PLND, allowing appropriate sensitivity analyses.

In conclusion, our lymphatic mapping study delineated a complex drainage pattern of the prostate that raises questions about whether targeted lymph node dissection can be implemented in PCa. Together with the low sensitivity for the detection of metastases, these results suggest that, for the time being, fluorescence SNL detection does not represent an alternative to a meticulously performed PLND in higher risk patients.

Author contributions: George N. Thalmann and Daniel P. Nguyen had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Nguyen, Thalmann.

Acquisition of data: Huber, Metzger, Schudel.

Analysis and interpretation of data: Nguyen, Huber.

Drafting of the manuscript: Nguyen.

Critical revision of the manuscript for important intellectual content: Huber, Metzger, Genitsch, Schudel, Thalmann.

Statistical analysis: Nguyen.

Obtaining funding: None.

Administrative, technical, or material support: None.

Supervision: Nguyen, Thalmann.

Other (specify): None.

Financial disclosures: George N. Thalmann and Daniel P. Nguyen certify that all conflicts of interest, including specific financial interests and relationships and affiliations relevant to the subject matter or materials discussed in the manuscript (eg, employment/ affiliation, grants or funding, consultancies, honoraria, stock ownership or options, expert testimony, royalties, or patents filed, received, or pending), are the following: None.

Funding/Support and role of the sponsor: The Fluobeam imaging device was provided at no charge by Fluoptics (Grenoble, France); however, the authors had complete control of the data and information submitted for publication.

The role of pelvic lymph node dissection (PLND) during radical prostatectomy (RP) remains a matter of continuous debate [1]. Sentinel lymph node (SLN) detection has been advanced as a potential alternative to PLND. In prostate cancer (PCa), the technique was first described using technetium Tc 99m bound to a colloid [2]; however, radioguided SLN detection has not come into widespread use. The use of the fluorescent dye indocyanine green (ICG) may open the door to broader acceptance of SLN techniques in PCa surgery [3]. Against this background, we provided a comprehensive description of lymphatic landing sites per anatomic region of the prostate using this SLN technique. We also evaluated the sensitivity of ICG-based fluorescence SLN detection to detect lymph node metastases in intermediate- and high-risk patients.

Detailed information on patient selection and detection technique is found in Supplement 1. From November 2012 through September 2015, 42 patients presenting with clinically localized intermediate- or high-risk PCa and scheduled for RP gave written informed consent to participate in our prospective study. ICG (Pulsion Medical Systems, Feldkirchen, Germany) was injected transrectally shortly prior to laparotomy. The first 12 patients enrolled received sextant injections (six injections into base, midportion, and apex of each prostatic lobe peripheral zone). The next 30 patients received injections into one of these sites: prostate base bilaterally, midportion bilaterally, apex bilaterally, left lobe (base, midportion, apex), or right lobe (base, midportion, apex). A near-infrared–sensitive probe (Fluobeam; Fluoptics, Grenoble, France) was used to collect fluorescence generated in the tissue under real-time image guidance. Independent of the findings of fluorescence SLN detection, an extended PLND was subsequently performed. An ex vivo fluorescence examination of all dissected lymph nodes was then carried out. Lymphatic landing sites per anatomic region of the prostate were depicted graphically. Diagnostic statistics assessed the value of ICG in detecting lymph node metastases.

Baseline characteristics of the patients are summarized in Supplementary Table 1. All 42 patients had one lymph node or more detected by fluorescence. The lymphatic mapping study showed the following results (Fig. 1; Supplementary Table 2): (1) The external and internal iliac regions encompassed the majority of SLNs; (2) the common iliac regions contained up to 22% of all SLNs; (3) a prostatic lobe can drain into the contralateral group of pelvic lymph nodes; (4) the fossa of Marcille also received significant drainage; and (5) practically all sites of the prostate can drain to different regions of the pelvis bilaterally.

Fig. 1

Percentages of sentinel nodes detected per drainage region with regard to anatomic sites of the prostate. The indocyanine green injection sites are depicted in the upper right corner. The orange zones represent the external iliac regions, the yellow zones represent the internal iliac regions, the green zones represent the common iliac regions, and the regions delineated by dashed lines represent the fossa of Marcille.

 

The drainage pattern did not show that distinct lymphatic pathways exist per prostatic anatomic region. Our results also underscored that lymphatics cross over to the opposite side and that the common iliac regions and the fossa of Marcille should not be overlooked during PLND, as combined, they may contain up to a third of all SLNs. The multitude of lymphatic landing sites as well as the individual variability of lymphatic drainage may represent obstacles to intraoperative SLN detection. In contrast to breast cancer, lymph node metastases in PCa do not follow a predefined pathway of metastatic spread, and there is no certainty that the histologic status of the SLN reflects the status of the entire pelvic node basin.

Among the 12 patients who received systematic ICG injections, a median of 15 SLNs per patient were removed (interquartile range [IQR]: 10–20). Five patients had a total of 29 lymph node metastases (Table 1). Of these, 16 were ICG positive, yielding a sensitivity for the detection of lymph node metastases of 55%. The negative predictive value was 95%, the specificity was 57%, and the predictive positive value was 8%. Repeating analyses with patients as the unit of analysis showed that four of five patients had ICG-positive lymph node metastases, yielding sensitivity of 80% for the correct staging of patients.

Hruby et al evaluated 38 patients who underwent laparoscopic RP [4]. The sensitivity of the SLN technique with ICG to detect metastases was 98% (42 nodes in 15 patients). Yuen et al reported on 66 patients who underwent RP and SNL detection using ICG [5]. Only nine metastases in six patients were detected, all of which were ICG positive. Our median number of nodes removed was 35 versus 18 in both previous reports. These findings suggest that our template leads to a notably more complete PLND, which would result in more precise sensitivity analyses. In addition, our median number of SLNs removed was 15 (IQR: 10–20), which is in the higher range of most previous studies using ICG alone or in combination with a radiocolloid [3], [4], [5], and [6]. Other authors may not have searched as thoroughly for SLNs in surgically less accessible regions such as the proximal part of the common iliac vessels, in the fossa of Marcille, and in the presacral regions. In the study by Yuen et al, only 2% and 1% of all SLNs were found in the common iliac and presacral regions, respectively [5].

The low sensitivity of ICG to detect metastases in the current study raises a concern about “skip” metastases. The presence of tumor cells in the lymph nodes may clog the feeding lymphatics and interfere with their ability to take up ICG, as has been postulated for radiocolloids [7]. Interestingly, we could sometimes visualize lymphatic drainage going around enlarged nodes. Going forward, emerging molecular imaging modalities using dyes conjugated with tumor-specific peptides may provide added sensitivity [8], [9], and [10]. It has to be noted that it is not possible to discriminate with certainty between primary landing sites and higher levels of drainage; however, the fact that fluorescent lymph nodes appeared 15–30 min after ICG injection and that direct drainage through lymphatic vessels could be seen in some cases suggests that these lymph nodes can be considered SLNs. Our study is limited by its small sample size. Nevertheless, it included a high number of lymph node metastases in the setting of extended PLND, allowing appropriate sensitivity analyses.

In conclusion, our lymphatic mapping study delineated a complex drainage pattern of the prostate that raises questions about whether targeted lymph node dissection can be implemented in PCa. Together with the low sensitivity for the detection of metastases, these results suggest that, for the time being, fluorescence SNL detection does not represent an alternative to a meticulously performed PLND in higher risk patients.

Author contributions: George N. Thalmann and Daniel P. Nguyen had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Nguyen, Thalmann.

Acquisition of data: Huber, Metzger, Schudel.

Analysis and interpretation of data: Nguyen, Huber.

Drafting of the manuscript: Nguyen.

Critical revision of the manuscript for important intellectual content: Huber, Metzger, Genitsch, Schudel, Thalmann.

Statistical analysis: Nguyen.

Obtaining funding: None.

Administrative, technical, or material support: None.

Supervision: Nguyen, Thalmann.

Other (specify): None.

Financial disclosures: George N. Thalmann and Daniel P. Nguyen certify that all conflicts of interest, including specific financial interests and relationships and affiliations relevant to the subject matter or materials discussed in the manuscript (eg, employment/ affiliation, grants or funding, consultancies, honoraria, stock ownership or options, expert testimony, royalties, or patents filed, received, or pending), are the following: None.

Funding/Support and role of the sponsor: The Fluobeam imaging device was provided at no charge by Fluoptics (Grenoble, France); however, the authors had complete control of the data and information submitted for publication.