Surgical Approach Does Not Impact Margin Status After Partial Nephrectomy for Large Renal Masses

Introduction

There has been an increasing utilization of partial nephrectomy (PN) in the management of renal tumors over the past decade. ¹ Cancer-specific outcomes between radical nephrectomy and PN have been shown to be comparable, with nephron preservation also conferring decreased risk of chronic renal insufficiency and initiation of dialysis with potentially improved overall and cardiovascular survival. ^{2 –5} For renal masses <4 cm, recent guidelines advocate that PN should be offered to all patients in whom an intervention is indicated and possess a tumor amenable to this approach. ^6,7 Moreover, several studies have demonstrated the oncological and functional efficacy of PN in the surgical management of larger renal masses. ^{8 –10}

The ultimate goal of PN is the complete removal of tumor, while attaining cancer-free margins. ¹¹ The effects of positive surgical margin (PSM) on oncologic outcomes after PNs remain unclear. Historically, margin status was not believed to impact oncologic outcomes, ^{12 –14} but several recent studies have suggested that PSMs play an important role in outcomes ^{15 –17} and may be influenced by surgical approach. ¹⁸

While the influence of PSMs on oncologic outcomes remains highly contested, margin status nevertheless serves as a quality indicator for oncological control in PN. ¹¹ Most of the available literature discussing PN and margin status, especially at the population level, has focused mainly on small renal masses. ^13,17,18 In this study, we sought to evaluate the relationship between PSMs and surgical approach in a national, contemporary cohort of patients undergoing PN for large (>4 cm) renal masses and to identify predictors of PSMs in this population, and secondarily to evaluate the impact of PSM on survival in this population.

Materials and Methods

Results

Between 2010 and 2013, 7495 patients underwent PN for pT_1b and pT_2a renal masses. Of these, 2813 (37.5%), 882 (11.8%), and 3800 (50.7%) underwent RAPN, LPN, and OPN, respectively. Patient demographics, hospital characteristics, and baseline clinicopathological features stratified by surgical approach can be found in Table 1. Overall, PSMs were observed in 504 (6.7%) cases. On multivariable analysis (Table 2), surgical approach (LPN or RAPN vs OPN) was not significantly associated with PSM (p = 0.12 and p = 0.44, respectively). Likewise, tumor stage (T_2a vs T_1b) showed no significant association (p = 0.18). Significant predictors of PSM include age >60 years (OR 1.57 [95% CI 1.01–2.44] p = 0.05), African-American race (OR 1.52 [95% CI 1.06–2.17] p = 0.02), education level (OR 1.48 [95% CI 1.03–2.14] p = 0.03), rural setting (OR 4.82 [95% CI 2.45–9.46] p < 0.001), PN at a nonacademic center (OR 1.57 [95% CI 1.15–2.15] p = 0.004), mixed histology (OR 1.84 [95% CI 1.04–3.24] p = 0.04), and undifferentiated tumor grade (OR 2.42 [95% CI 1.26- 4.65] p < 0.01).

Upon running a subsequent multivariable analysis on the newly identified, clinically staged cohort of 6555 patients from the sensitivity analysis (Appendix Table A1), surgical approach (LPN or RAPN vs OPN) was not significantly associated with PSM (p = 0.28 and p = 0.54, respectively), nor was tumor stage (T_2a vs T_1b, p = 0.78) (Table 3). Moreover, conversion-to-open surgery showed no significance (CPN vs OPN, p = 0.98) (Table 4).

After excluding 3470 patients from the OS analysis attributable to being diagnosed in 2013 or for having more than one malignancy, 263 patients (6.5%) had a PSM within the cohort of 4025 patients (Appendix Table A2). In the unmatched cohort, significant differences existed in terms of margin status by surgical approach (p = 0.02), patient age (p = 0.002), insurance type (p = 0.01), center volume (p < 0.001), facility type (p < 0.001), tumor histology (p = 0.002), and grade (p = 0.02). However, after 1:1 propensity score matching, the 243 (matching rate: 92.4%) patients with a PSM to patients with negative surgical margin, the groups were well balanced, and none of the aforementioned variables retained their significance. Kaplan-Meier curves illustrating OS are shown in Figure 1. Margin status was not significantly associated with OS in the propensity score-matched cohort (HR 0.95 [95% CI 0.47–1.92] p = 0.88).

OPEN IN VIEWER

FIG. 1.

Kaplan-Meier estimates of overall survival based on margin status.

Discussion

There has been a rapid adoption and increasing use of minimally invasive surgical techniques in urologic surgery, particularly for renal surgery and PN. ¹ Whether surgical modality is related to PSMs after PNs remains controversial, as there are contradicting studies both demonstrating and refuting the relationship. ^{13,17 –19} Using contemporary data from the NCDB, we demonstrate that surgical approach is not independently associated with increased risk of PSMs in patients undergoing PN for large (>4 cm), cT_1b and cT_2a masses. This is in contrast to the recent study by Tabayoyong and colleagues, which looked at patients undergoing PN for clinical T_1a RCC and concluded that minimally invasive surgical approaches were associated with higher rates of PSMs. ¹⁸ In support of our findings, Ficarra and coworkers demonstrated in a large retrospective matched-pair analysis that the rate of PSM after an OPN was comparable to that of RAPN, 5.5% and 5.7%, respectively. ²⁰ Similarly, Porpiglia and colleagues found no significant differences in rates of PSM as dictated by surgical approach when looking at data from patients undergoing PN for cT_1b tumors. ¹⁹

Another important finding was that PN performed at an academic facility was independently associated with lower rates of PSMs. Since we were unable to account for individual surgeon characteristics or volume, the reason behind the protective effect against PSM afforded by an academic center is speculative. However, it is possible that a more experienced cohort of surgeons is operating in these settings. As a result, these surgeons may be better equipped to handle the greater technical difficulty of performing a PN in large masses, resulting in a different rate of observed PSM. The implication from this hypothesis makes a compelling argument for the consolidation of care at academic centers, especially for the more technically challenging cases, thereby allowing for better quality of oncologic care.

In addition, we found that increasing tumor size (T_1b vs T_2a) did not confer increased risk of having a PSM on multivariable analysis. Although current guidelines advocate for the management of T₁ masses with PN when amenable, ^6,7 several studies have demonstrated the technical feasibility, oncologic efficacy, and safety for performing PN on larger tumors. ^{8 –10} A study by Benoit and coworkers showed that PN on tumors >7 cm had similar outcomes to tumors 4 to 7 cm with respect to Trifecta attainment (a combination of negative surgical margins, functional preservation, and complication-free recovery). ²¹ Conversely, in a recent population-based report using data from the Ontario Cancer Registry, it was found that tumor stage was significantly associated with PSM when comparing pT_1b to pT_1a masses. ¹³ However, no significant association was found when looking solely at tumor size per 1-cm increments. ¹³ Our study failed to find any relationship between categorical tumor size and rates of PSM, suggesting that size alone should not alter surgical approach, and should rather be dictated by surgeon comfort level with an individual tumor.

Currently, the relationship between PSM and age at the time of surgery remains unclear. Findings from our analysis indicate that increasing age beyond 60 was associated with an increased risk for PSM, which is in contrast to findings demonstrated in a recent study by Ani and coworkers. ¹³ Schiavina and colleagues hypothesize that the independent predictive value of age at the time of surgery for PSM after a PN largely revolves around the idea of increased renal and cardiovascular comorbidities that most commonly exist in older patients. ²² As renal function deterioration and rate of end-stage renal disease correlate with increasing age, ^23,24 urologists may be more inclined to preserve as much parenchyma as possible, which as a result may lead to an increased risk of PSM. ²² Furthermore, the age-associated anesthesia mortality risk with elderly patients ²⁵ could serve as an impetus for a more expeditious surgery.

While the significance of obtaining negative surgical margins as a surrogate for complete tumor removal has been shown in studies involving bladder, prostate, and breast cancer, ^{26 –28} the consequences are less clearly defined for renal masses and remain controversial. ^{12 –17} The main limitations of the various studies looking at the oncologic impact of PSMs in PNs are the short- to intermediate-term follow-up and small sample sizes. ¹⁷ It has also been argued that the clinical relevance of PSM after a PN stems largely from the intrinsic histopathological properties of the primary renal lesion with regard to the tumor grade and stage. ¹⁶ In this study, undifferentiated tumor grade was significantly associated with PSM. However, margin status did not affect OS (p = 0.88). This is not surprising, given our short follow-up period, which was dictated by the fact that 2010 marked the first year that data regarding surgical approach were available in NCDB. Longer follow-up is needed to fully understand the survival implication of PSMs, especially in larger or more complex masses.

The findings of our study must be interpreted within the context of the study design. The retrospective nature of the study and lack of standardization of surgical modalities is a limitation. We could not account for individual surgeon volume, and instead used center volume. This may be a reason for why center volume was not predictive of PSM, but the type of center (academic vs nonacademic) was. Furthermore, because of limitation of the database, we were unable to account for cancer-specific or recurrence-free survival. We were also not able to account for tumor complexity such as nephrometry score ²⁹ ; however, with our analysis, we were able maintain a relatively homogenous cohort of patients with large, noninvasive tumors. Finally, we were unable to account for margin length, which some have speculated to have important implications.

Conclusions

In a contemporary national cohort, surgical approach was not associated with increased risk of PSM for large, noninvasive renal masses. Furthermore, increased size from T_1b to T_2a was also not associated with increased risk. These data suggest that surgical approach should be dictated by surgeon comfort level with an individual tumor, and not by size criteria alone.