Safety and Efficacy of Robot-assisted Radical Prostatectomy in a Low-volume Center: A 6-year Single-surgeon Experience

Discussion

Currently, several large series (2-7, 13-30) confirm that RARP is an effective treatment for clinically localised prostate cancer, if performed in high-volume referral institutions.

However, these results may not be applicable to patients treated at low-volume centres. Based on these considerations, we sought to evaluate the safety and efficacy of RARP in our low-volume institution.

The primary goal of RARP is the complete removal of tumour. In this regard, PSM status is considered as a predictor of cancer control. In a recent review on RARP outcomes from high-volume centres, Patel et al. (6) reported an overall PSM rate of 15.7%. In our experience, overall PSM rate was 13.2%, particularly 9.8% in pT2 and 25.6% in pT3 cases. Accordingly, our local cancer control rate is in line with those from referral institutions (4-6, 13-19). In fact, they are consistent with those from large RARP series, such as from Badani et al. (13) showing an overall PSM rate of 13% in 2,766 RARP patients, with 13% for pT2 and 35% for pT3 disease. Recently, in 184 patients, Suardi et al. (15) reported a 15.7% PSM rate, with 2.5% for pT2, 34% pT3a and 50% for pT3b patients, respectively. Higher PSM rates (for all pT, pT2 and pT3 cases) were observed in other high-volume reports by Ficarra et al. (16) (29.5%, 10.6% and 59%, respectively), Galfano et al. (17) (26.0%, 15.0% and 45.0%, respectively) and Sooriakumaran et al. (18) (21.6%, 16.0% and 59.1%, respectively), whereas excellent results were shown by Coelho et al. (19) (10.6%, 5% and 27.5%, respectively). Although PSM occurrence is highly variable (6.5% to 32%) after RARP (4), these findings suggest that RARP offers good results regarding PSM status both in low- and high-volume centres.

Overall, RARP has shown to provide effective BCR-FS rates also at a long-term follow-up (13-15, 18, 20, 21). In one of the largest RARP series published in the literature on 2,766 patients, Badani et al. (13) noticed 95 (7.3%) cases with BCR and a 5-year BCR-FS rate of 84%. In another large cohort of 944 patients, Sooriakumaran et al. (18) reported a BCR-FS rate of 84.8%: the 5-, 7- and 9-year BCR-FS rates were 87.1%, 84.5% and 82.6%, respectively. Recently, Diaz et al. (21) followed 483 men after RARP at a high-volume tertiary centre, showing a 10-year BCR-FS rate of 73.1%.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1.

Biochemical recurrence (BCR)-free survival curves for (A) preoperativePSA level, (B) D'Amico risk, (C) pathologic stage, (D) pathologic Gleason score, (E) pathologic nodal status and (F) surgical margin status.

When analyzing BCR in our cohort, cumulative BCR-FS rate was 79.0%, with1-, 3- and 6-year BCR-FS rates being 85.7%, 77.5% and 53.9%, respectively. Overall, our BCR rates appear to be higher than those from high-volume centres (13-15, 18, 21). However, since BCR is strongly predicted by factors, such as D'Amico risk groups, as well as pathological Gleason score, stage and margin status (18, 20, 21), accurate comparison among series must take these predictors into account. Indeed, compared to reports from referral institutions, our cohort shows higher baseline PSA levels and includes higher rates of intermediate- and high-risk cancers (75.6%). Furthermore, the proportions of cases with pathological Gleason score 7-10 (93.2%), as well as positive nodes (14%), appear to be higher in our cohort. In sum, these selection biases have to be considered. In addition, our BCR-FS rates have to be interpreted with caution considering the high number of patients lost at 4- and 6-year follow-up. This might be attributed to the Swiss health care system where most patients receive late follow-up with local urologists or general practitioners. This, again, can lead to case selection over time. On the whole, the abovementioned factors might have adversely affected BCR-FS in our cohort.

• Download figure
• Open in new tab
• Download powerpoint

Figure 2.

Kaplan-Meier plot depicting continence (A) and potency (B) recovery after robot-assisted radical prostatectomy (RARP).

Preservation of functional outcomes is fundamental after RARP due to great impact on patient satisfaction and quality of life. In the literature, few studies report on the long-term functional outcomes after RARP (6, 17, 20, 22). In our cohort, 98.4% of patients were fully continent after surgery, with the 1-, 3- and 6-year continence rates being 98.4%, 98.5% and 100%, respectively. Overall, our results are equivalent to those from high-volume centres, showing mean (range) for continence rate of 91.8% (70-97%) at 12 months (6, 17, 20, 22). Thus, our data are consistent with results from large series, such as Menon et al.'s (22) showing a continence rate of 95.2% at 1-year follow-up. In another paper by Ficarra et al. (20) on 183 patients, 146 (79.8%) were fully continent at follow-up with rates of 73% at 3 months, 87% at 6 months and 91% at 1 year, respectively. Encouraging outcomes were also reported by Galfano et al. (17) with no patient using safety liners at 1 year after RARP.

Potency recovery after radical prostatectomy is influenced by numerous factors, including baseline patient characteristics, nerve-sparing techniques, definition of potency and methods used to collect data (3). In our cohort, overall potency rate was 68.2% with a median (IQR) time to potency of 3 (3-4) months. In the literature, RARP series from referral centres show mean (range) values for potency rates of 70% (54-90%) and 79% (63-94%) at 1- and 2-year follow-up, respectively (3). Notably, excellent results were reported by Menon et al. (23) with a potency rate of 94% at 18 months. Similar rates were observed also by Patel et al. (24).

The ‘best case scenario’ outcomes after RARP are represented by the ‘trifecta’ (25), including the concurrent achievement of continence, potency and cancer control. Trifecta results after RARP were first reported by Shikanov et al. (26) who showed rates of 16%, 44% and 44% at 3, 12 and 24 months, respectively. Subsequently, Novara et al. (27) evaluated 242 consecutive patients after RARP and a trifecta outcome was achieved by 137 (57%) patients at 12 months. Excellent results were observed by Patel et al. (28) with 65.3%, 86% and 91% at 3, 12 and 18 months, respectively. Particularly, the outcomes during the learning curve were not included in this study. More recently, Xylinas et al. (29) evaluated 500 consecutive patients and a trifecta outcome was achieved in 44% and 53% of men at 12 and 24 months, respectively. In the present series, trifecta rate approached 55% at 6 years after RARP. However, trifecta score is greatly influenced by tumour and patient characteristics, such as preoperative potency status, and surgical experience. In addition, in the present analysis, we included all cases of the surgical learning curve. The mentioned factors might have influenced trifecta outcomes in our series.

Safety profile represents a major outcome after RARP. Previous studies have shown conflicting results, with complication rates ranging from 3% to 26% (5). Indeed, in the literature, several limitations are associated to complication reporting. Specifically, lack of standardisation for defining, different methods of collection and incomplete reporting of data render this outcome a difficult one to be assessed and compared among series. Applying the Modified Clavien System (12), we identified complications in 27% of patients. Our complication rate appears to be higher than those reported by high-volume institutions (5, 19, 22, 30). However, such a difference might be partly attributed to the strict methodology of data collection in our study. As mentioned above, first, all patients of the learning curve were included. Second, we assessed complications not only during the intra-operative and early post-operative phase but also over the entire duration of follow-up. Third, any deviation from the perioperative standard was classified as a complication, including clinically insignificant events, such as asymptomatic lymphoceles requiring no treatment. Notably, those represented the most prevalent complications (27/125; 21.6%).

The present study has some limitations. Although data were collected prospectively, the analysis is retrospective and, thus, subject to the inherent limitations of retrospective analyses. Additionally, the high rate of patients with more aggressive diseases has to be taken into account when analysing oncological outcomes in our cohort. Follow-up might be impaired by selection bias, attributed to established processes within the Swiss health care system. Most importantly, all our results have to be considered from the perspective that all patients of the learning curve were included. Strengths of our study include the solid data acquisition, using validated tools to assess functional outcomes, as well as a strict reporting of complications. Moreover, all procedures were performed by a single experienced surgeon. Nevertheless, this study confirms that also centres with limited caseload can offer RARP with good oncological and functional results.

In conclusion, in our low-volume centre RARP confirmed to be a valuable option for the treatment of clinically localised prostate cancer. Although all cases of the learning curve were included and we operated mostly on intermediate- to high-risk patients, the results of the present study are consistent with those from high-volume referral centres. Therefore, RARP proved to be an excellent treatment option with favourable oncological and functional outcomes also in a low-volume institution.