Our study reports the outcome of DSNB in a contemporary era of liberal cross-sectional imaging. 30.3% of all patients had lymph node metastasis. We report a 0% FNR for DSNB at a mean and minimum follow-up of 65 and 24 months, respectively. There is no difference in mean cancer-specific survival between DSNB and ILND patients (85 vs 92 months, p = 0.124, log rank test) with lymph node-positive disease. To our knowledge, this is the first study from a large tertiary referral centre to report the survival and diagnostic performance on all patients who have had lymph node surgery (i.e., DSNB and ILND) with curative intent in penile cancer.
Five studies have reported results of DSNB in more than 100 patients with adequate follow-up (6–8, 10, 12). Their DSNB lymph node positive rates were quite variable, from 4.9–22.3% of patients. Our result of 11.2% is in line with these studies. To our knowledge, published series on DSNB do not report the outcome of cN + patients, which is essential to ascertain the validity of such protocol (6–9). For example, a protocol with 5% FNR might have had lower cN + pathological positive rate in ILND as opposed to protocol with 10% FNR. Ours is the first paper to report the outcomes from both cN + and cN0 patients, to provide a comprehensive strategy to manage all penile cancer patients and it can act as a benchmark for future studies. The Amsterdam approach (12) has used SPECT CT to circumvent the difficulties in using US FNAC alone for imaging. However, in our experience we find that the resolution of CT in SPECT imaging is inferior to conventional CT or MRI.
DSNB had been established as a gold standard in management of cN0 disease as ILND, as a diagnostic test, has unacceptable short- and long-term complication rates. Most studies report a false-negative rate of DSNB between 5–13%, having used clinical examination of palpability to stratify patients into cN0 and cN + disease. However, it suffers from considerable inter-observer variation and is particularly unreliable in obese patients. We also know SN localisation does not work well in the presence of enlarged nodes full of tumour or if there is infection. Subjecting the latter group of patients after negative US FNAC to DSNB may not be reliable. Hence, we had a different approach and assessed the stratification of patients into cN0 and cN + disease groups using cross-sectional imaging and proceeding to up-front VEILND (Video endoscopic Inguinal lymph node dissection), if equivocal, to address this issue. Combined clinical and radiological stratification identifies abnormal lymph nodes, as opposed to enlarged nodes. Hence, we believe that this stratification approach helped us to reduce the false-negative rate substantially with acceptable higher ILND rate.
Open ILND had been found to have a high short- and long-term complication rate of 30–70%, even in the contemporary era (13, 14). The fear of an unacceptable complication rate has forced clinicians to resort to DSNB as a diagnostic tool in borderline cases of node enlargement. However, the advent of newer techniques, like VEILND and RAILND (Robotic assisted endoscopic inguinal lymph node dissection), with substantially reduced complication rates, has made ILND a viable alternative for DSNB in borderline cases. 22/25 cN + patients in this series had VEILND and the results have already been published (11). Hence, our newer approach of clinical-radiological stratification and VEILND for borderline cases has offered the best approach to deal with lymph node metastasis for penile cancer patients.
In our series, DSNB is a highly reliable test with sensitivity of 100%. Previous study has explored novel technique of combining DSNB with FDG PET scanning to reduce false-negative rate and the authors reported a FNR of 5.6% and 100% mortality in those 2 FN patients (15). Hence there is a quest to improve false negative rate in penile cancer lymph node staging. Centralisation of penile cancer services with an individualised medicine approach over a protocol-based approach and the judicious use of available radiological imaging will help to improve the outcome in this disadvantaged group of patients.
Our study has some limitations. This is a non-randomised series retrospective analysis from a single centre where most decisions are made by single supra-network MDT. Hence the stratification methodology needs to be validated in other centres. Our centre also offers VEILND/ RAILND to all the patients who require inguinal lymph node dissection with considerably reduced morbidity. As minimally invasive inguinal lymph node dissection may not be available in all centres, stratification of patients to cN + may lead to higher ILND rates and higher morbidity in those centres. Hence, subjecting patients to highly morbid open ILND to reduce false-negative rate of 10% needs to be approached with caution.
We performed DSNB on all patients with invasive penile cancer including low risk disease for multiple reasons. Our regional unpublished data prior to this study showed 8% LN metastasis in low-risk disease. Jakobsen et al. also showed that 6% of their low-risk cohort had lymph node-positive disease (7). Further, we performed DSNB at the time of primary surgery as a standard to avoid multiple general anaesthetics and hospital visits. A significant proportion of these patients have not had a biopsy as the cancer was diagnosed on clinical grounds and it is difficult to be certain about the grade and stage of disease at presentation for risk stratification. In addition, 5 of the 10 low-risk patients had T1 disease at incisional biopsy and large tumours. Eventually they turned out to be G1pT1 disease though there was a clinical prediction for it to be intermediate or even high-risk, disease. None of our 10 patients suffered additional complications due to DSNB.
As previous studies have shown that most recurrences occur within 24 months, we consider this follow up of 24–150 months is adequate to detect any false-negative cases. We have also reported the harder end point of cancer-specific survival (CSS) which is similar between DSNB and ILND patients with node-positive disease.