Positive surgical margin following radical nephrectomy is an independent predictor of local recurrence and disease-specific survival

Positive surgical margins (PSM) are uniformly considered an adverse outcome associated with incomplete cancer removal and are often allied with increased risk of local or distant recurrence [1‐3].

The management of patients with PSM remains a challenge in renal cell carcinoma (RCC) patients, with controversy persisting over the need for more rigorous follow-up or for immediate adjunctive therapy. Over the past years, many studies have investigated the outcomes of patients reported to have a PSM following partial nephrectomy (PN). However, a consensus has yet to be reached on the prognostic significance and optimal management of these patients, in comparison to patients with negative surgical margins [4, 5]. During the last decade, PN has become the standard of care for small exophytic, and favourably located renal tumours, based on its provision of equivalent cancer control and better preservation of long-term renal function. However, radical nephrectomy (RN) remains a commonly performed surgery, primarily in the cases in which partial resection is not feasible due to unfavourable tumour location or in patients with locally advanced tumour growth [6]. The data regarding the oncological outcome for PSM following radical nephrectomy is limited.

In the current study, we examined the potential predictors of PSM following RN and investigated long-term oncological outcomes in cases of PSM.

Data included all patients who underwent elective radical nephrectomy for renal masses between 1988 and 2013. Patient demographics and surgical details were collected retrospectively following an approval given by our Institutional Review Board. Informed consent was impossible or impracticable to obtain for such research. Research was done only after consideration and approval of a research ethics committee.

Clinical variables recorded included age, gender and co-morbidities. Tumour-related variables included tumour size, side and multifocality. Surgical variables included type of operation (i.e. open or laparoscopic). Pathological variables included capsular invasion, vascular invasion (including renal vein and inferior vena cava), renal pelvis invasion, perinephric fat extension and tumour necrosis. Tumour stage was coded as a dichotomous variable, pT2 or lower vs. pT3. Data pertaining to tumour location or percent parenchyma involvement were unavailable for the majority of patients and were not included.

Patients with benign histology (including metanephric adenomas, angiomyolipomas and oncocytomas) were excluded from this analysis. Additional exclusion criteria included patients with malignant tumours other than renal cell carcinoma (urothelial cell carcinoma, sarcoma, neuroendocrine tumour, squamous cell carcinoma, leiomyoma and liposarcoma) or metastatic disease upon diagnosis (Fig. 1). Surgical margin status from the radical nephrectomy was recorded as positive or negative based on macroscopic and microscopic examination of the radical nephrectomy specimen. In all cases, the renal vein, renal vein margin and all other margins were examined grossly and sampled for microscopic examination if suspected of involvement. The interface between tumour and perinephric fat was sampled in all cases to evaluate perinephric fat invasion [7]. PSM were identified from pathology reports (determined by specialized pathologists), using the standard pathology criteria that define a PSM by either extension of tumour to the inked surface of the resected specimen on final pathology or evidence of tumour thrombus invasion into the segmental venous branch, renal vein or inferior vena cava. Tumour size was determined by measuring the maximal diameter of the tumour at pathological examination. Follow-up was conducted according to the standard clinical practice at our institution. In general, follow-up consisted of physical examination, chest radiographs and kidney imaging every 6 to 12 months during the first 5 years and annually thereafter. Patients with PSM on final pathology were observed at similar intervals with serial imaging. None of the patients was treated with immediate adjuvant therapy. Metastatic progression was defined as unequivocal imaging findings indicative of distant organ involvement with or without a confirmatory diagnostic biopsy (based on the discretion of a multi-disciplinary team).

This study included 714 patients. The clinicopathologic demographics for these patients are provided in Table 1. Of 714 patients, 17 (2.4%) had positive surgical margins at RN. Univariate analysis revealed several variables potentially associated with PSM including tumour size (P = 0.001), advanced pathologic stage (P = 0.001), central location (P = 0.01), tumour necrosis (P = 0.001) and capsular invasion (P = 0.002). Histologic RCC subtype and operative method (laparoscopic vs. open) were not associated with PSM (Table 2). All variables found significantly related to PSM in the univariate analysis were introduced into a multivariable logistic regression analysis. On multivariate analysis, tumour size (P = 0.001), presence of necrosis (P = 0.002) and advanced tumour stage (P = 0.002) remained significantly associated with PSM (Table 3).

The achievement of negative surgical margins has a paramount importance and remains the main goal during oncologic surgery. PSM is recognized as an adverse prognostic sign for disease recurrence, especially in tumours of high malignant potential. The potential unfavourable effect of PSM on disease progression and CSS has been previously described in different types of tumours including prostate, rectal and breast cancer but not for renal cancer [8‐12].

In recent years, large series have analyzed the preoperative risk factors for PSM after PN. Several factors have been advocated as predictors, such as older age, tumour size, location, pathological stage, Fuhrman grade and indication (elective vs. imperative) [5, 13‐15]. Correspondingly, conflicting evidence has accumulated on whether PSM is also a significant risk factor for disease progression, while large-scale studies like Yossepowitch et al., Lopez-Costea et al. or Antic et al. suggest no correlation between PSM and local recurrence or distant progression [14, 16, 17]. Other investigators such as Kwon et al. and Bernhard et al. [18, 19] identified PSM as an independent risk factor for tumour recurrence in the setting of partial nephrectomy, though CSS and OS were not affected by margin status. Yet, despite this discordance, the main implication regarding disease recurrence rate for PSM in partial nephrectomy is the postoperative management. In lieu of the adverse outcome associated with PSM in other solid tumours, patients with PSM should be offered all therapeutic options including radical nephrectomy, repeat PN, energy ablation and vigilant observation [4, 20‐22].

Considering the conclusions outlined, it indicates that although the prognostic impact of PSM after PN is ambiguous, these patients should still be closely monitored. Hence, the arising question is whether this conclusion can be applied for patients who undergo radical nephrectomy. To date, PN has been considered the standard of care for the treatment of most renal tumours, with long-term oncologic results equivalent to those of RN. However, PN is unsuitable in some patients with localized RCC due to locally advanced tumour growth or unfeasible because of unfavourable tumour location or significant deterioration in patient health [6]. Thus, the predictive factors for PSM following RN and its implication on patients’ follow-up are a matter of interest. In the current study, positive margins were associated with pathologic stage (≤ pT2 vs. pT3), tumour size and tumour necrosis. Increased pathological stage is associated with positive surgical margin in other malignancies treated surgically such as prostate, bladder and oral cavity cancer [23‐25]. According to the current paper, we have demonstrated the association between locally advanced disease (pathological stage (≤ pT2 vs. pT3)) and the presence of PSM. In addition, vascular involvement and tumour necrosis are associated with adverse oncology outcomes in renal cancer and therefore were found to be associated with PSM in addition to pathological stage by multivariate analysis.

In regard to oncologic outcome, previous studies have reported different factors to predict survival after RN, including tumour stage, size, grade and necrosis [7]. An earlier study by Leibovich et al. [26] has reviewed the prediction of progression after RN for patients with clear cell RCC. In his study, Leibovich and colleagues presented a rate of 0.7% PSM. Univariate analysis revealed significant association between PSM and metastasis, yet this factor did not remain significant in a multivariate modeling. Frank et al.[7] presented similar results, including the rate of PSM (0.8%). However, given the small number of cases, Frank and colleagues chose to exclude PSM from further analysis. In the current study, PSM is documented in 17 cases (2.4%). These differences in PSM proportions should be attributed to the pre-disposing tumour properties. Although tumour size was comparable between the studies, nearly 40% of the patients in the current study had pathologic stage of T3 and more, in comparison to 36 and 31% presented by Frank et al. and Leibovich et al., respectively. Moreover, no data is available on the rate of vascular invasion, which is also found to be a significant risk factor for PSM. Finally, both studies conducted analysis only on patients with clear cell RCC, which may have created a selection bias [14].

Our study showed, using multivariable analysis, that the only factor that could predict local recurrence was PSM. In parallel with Leibovich et al., PSM was not associated with metastatic rate, yet tumour necrosis, size and stage could predict metastatic progression. PSM was also associated with CSS along with tumour necrosis and stage, yet not with OS. Further analysis revealed that the effect on survival is so profound that patients with PSM have similar median survival rate to those with metastatic disease upon diagnosis. These results suggest that patients with positive SM, particularly with advanced tumour stage, should be offered more appropriate postoperative surveillance programs, including close monitoring. Currently, we cannot address the role of adjuvant therapy in order to improve the rate of local recurrence, metastasis or cancer-specific survival, due to lack of data to support it. In future, studies that will address the role of adjuvant therapy for high-risk renal cancer following radical nephrectomy should be stratified according to the status of SM.

Our study is not without limitations. First, this study is retrospective, with the entire attendant imprecision associated with the large recollection of data. Second, the number of patients with PSM is small. We acknowledge the fact that 17 patients is a small number; however, giving the potential effect on patients’ outcome and prognosis, analysis of this group, despite its small number, is rather important. To date, the literature relating to this subject is limited, specifically in regard to the association between PSM (following RN) and prognosis. To our knowledge, the current study is the first to directly investigate this association and suggest on its clinical implications. Despite the relatively small number and its limitations, being a retrospective study, we believe that such findings would encourage future randomized clinical trials and will shed the necessary focus on this important entity.

In this study, we defined the clinical significance of a PSM after RN for RCC. Resection margin appears to be an independent predictor of local recurrence-free and disease-specific survival for all patient subsets. Knowledge of the risk factors for PSM and its influence on disease progression may help clinicians to assess the effects of tumour characteristics on the oncological outcomes following radical nephrectomy, which can be used to improve treatment. The results of this study underscore the significance of histologic resection margin as a prognostic factor after RN.

Future randomized clinical trials are required before we could confirm that the higher incidence rate of PSM does translate into a poorer survival rate.