What’s known on the subject? and What does the study add?

It is known that the clinical and oncological effects of positive surgical margins (PSM) after partial nephrectomy (PN) have been addressed previously. In recent years, studies which showed that PSM does not have oncologically worse effects have increased. However, data regarding long-term results of PSM is scarce. Results of our study indicated that PSM after PN does not have clinically and oncologically worse effects in long-term follow-up.

Introduction

The percentage of incidental renal mass detection has increased due to the increasing frequency of diagnostic imaging methods. Nephron-sparing surgery is currently recommended for patients with organ-confined renal cell cancer (RCC). In this way, the protection of kidney functions is prioritised. Partial nephrectomy (PN) is the preferred treatment for organ-confined renal masses with equivalent oncological and superior functional outcomes compared to radical nephrectomy (1,2). As a result of the advancement of technology and growing clinical knowledge, PN may be involved in more challenging cases.

As a result of insufficient tumour resection or persistent microscopic tumour extension, which may increase in difficult cases, positive surgical margins (PSM) appear in the histopathological evaluation (3-5). However, several studies concluded that the survival of PSM patients was not worse (3-6). In certain studies, local recurrence was reported as a result of aggressive tumours, and course of the disease could be worse. These patients had high-stage tumours with higher Fuhrman grade at the time of diagnosis (3,7).

The aim of the current study was to evaluate the oncological effects of microscopic PSM in histopathology of patients who underwent PN for clinically localised RCC and disease management.

Materials and Methods

Following the approval by the local ethics committee (approval date: 20.01.2020, decision number: 80/08), the records of patients who underwent PN for renal mass between 2006 and 2018 were reviewed retrospectively. Furthermore, this study was a retrospective analysis of the database, covering all clinical, surgical, oncological and follow-up data for more than 400 consecutive patients who underwent open or laparoscopic PN in our clinic. Demographic data, peri-operative characteristics and histopathological and follow-up outcomes of patients were recorded. Computed tomography (CT) and/or magnetic resonance imaging (MRI) were used for pre-operative renal and tumour imaging, and thoracic X-ray or CT data were recorded.

Tumour size, localisation, clinical stage, surgical method and approach preferences were also recorded. All renal scoring systems were calculated by the same urologist. Additionally, tumour size was measured as the longest diameter of the tumour. Procedures were carried out by a team of four experienced surgeons with at least 10 years of urooncological experience. In the event of suspicion of macroscopic PSM, the resection was extended to the parenchyma and the procedure was completed. Histopathological assessments were conducted by a pathologist with 18 years of experience. Moreover, microscopic PSM was identified as the entity of cancer cells on the inked surface of the specimen.

Peri-operative complications were evaluated according to the modified Clavien-Dindo classification (8). Tumour subtype, Fuhrman nuclear grade (9), pathological stage and other histopathological features were also recorded. Hence, the clinical follow-up scheme after PN consisted of clinical visits every 3 months for the first year. In addition, all patients were examined at regular periods following PN with serum creatinine, liver function tests and thoracic and abdominal contrast-enhanced CT or MRI at 6 months post-operatively and every 12-24 months thereafter. In PSM patients, in addition to standard cross-sectional imaging, ultrasonography was performed every 6 months for the first 3 years and active surveillance was performed. Patient data were also collected from follow-up cards that were filled in at each admission and from patient interviews. Masses with benign pathology (n=46), non-RCC malignant masses (n=11) and patients with missing data (n=77) were excluded from this study. As a result of histopathological evaluation, patients with PSM (group 1) and patients with negative surgical margin (NSM) (group 2) were compared.

Statistical Analysis

One-sample Kolmogorov-Smirnov test was used to verify if the data displayed a normal distribution for numerical variables. Mean ± standard deviation was found in the data with normal distribution, and median interquartile range (IQR) values were recorded in the data with no normal distribution. Numerical variables were compared to the Student’s t-test when parametric test criteria were found. In the absence of such criteria, Mann-Whitney U test was used. Moreover, two proportion z-tests, Pearson’s chi-squared test and Fisher’s Exact test were used to evaluate if there was a discrepancy between the percentages of categorical variables. Binary logistic regression analysis was used to obtain independent risk factors relevant to PSM. Kaplan-Meier survival curves with a 95% confidence interval (CI) were used to assess the effect of surgical margin status on overall survival (OS), cancer-specific survival (CSS) and recurrence-free survival (RFS) after PN. The logrank test was used to compare survival results between PSM and NSM patients. In addition, multivariate Cox regression analysis was used for variables that were statistically significant in the univariate analysis and used to evaluate the factors influencing survival, and CI was given with hazard ratio. For all tests, the probability of first type error was a=0.05. Statistical analysis of the study was carried out using IBM SPSS 22.0 package programme.

Results

A total of 302 patients who underwent PN due to RCC were analysed in our clinic. PSM was observed in 38 (12.6%) patients. No macroscopic PSMs or residual tissues were present in all cohorts. Cases with PSM were specified as a microscopic entity in the parenchymal resection margin. There was no difference in demographic characteristics between the two groups. The tumour size measured by radiological imaging was 38.6±15.2 mm in group 1 and 39.0±17.4 mm in group 2. The Padua score was 8.3±2.2, and the C-index value was 1.8±0.5 in group 1 and 8.3±1.6 and 2.1±0.9 in group 2 (p=0.922 and p=0.016, respectively). Open PN patients were 76.3% in group 1 and 67.8% in group 2. Retroperitoneal approach was favoured in the majority of patients and 65.8% and 75.8%, respectively, in groups 1 and 2 (p=0.187). Non-ischaemic procedures were the majority in group 1 with a rate of 60.5% (p<0.001). Histopathological outcomes in both groups were not statistically different. Peri-operative complications in both groups were frequently low grade (p=0.249) (Table 1).

Multivariate binary logistic regression analysis model adjusting for covariates stated by univariate analysis showed that RENAL nephrometry score (OR: 1.438, 95% CI: 1.202-1.850, p=0.037) and C-index value (OR: 0.224, 95% CI: 0.070-0.723, p=0.012) were significant predictive factors for PSM. The median follow-up period was 85.2 (IQR, 10.1-160.4) months in group 1 and 83.7 (IQR, 13.5-153.9) months in group 2 (p=0.869). Three patients (7.9%) in group 1 and nine (3.4%) in group 2 underwent radical nephrectomy due to local or systemic recurrence at a median period of 28.2 months from PN (p=0.181). The remaining 35 patients in group 1 underwent intensive surveillance. Three (7.9%) patients in group 1 and 14 (5.3%) in group 2 died of cancer-related condition (Table 2). Furthermore, the 5-, 10- and 15-year OS rates were 93.6%, 76.5% and 51.0% in group 1 and 90.0%, 71.7% and 57.9% in group 2, respectively (p=0.580, by logrank test, Figure 1A). CSS did not vary, with 96.6%, 90.1% and 60.1% for 5-, 10- and 15-year CSS rates for PSM patients compared to 95.6%, 87.9% and 79.9% for those with NSM, respectively (p=0.948, by logrank test, Figure 1B). Further, 15-year RFS was 91.2% in group 1 and 94.3% in group 2 (p=0.332, by logrank test, Figure 1C).

Based on a multivariate Cox regression analysis, Charlson Comorbidity index (CCI) (HR: 1.343, 95% CI: 1.163-1.551, p<0.001) and high tumour stage (HR: 3.886, 95% CI: 1.576-9.580, p=0.003) were independently predictive of OS (Table 3). We also observed that CSS was independently decreased with mid-renal tumours (HR: 4.157, 95% CI: 1.478-11.692, p=0.007), high tumour stage (HR: 6.274, 95%CI: 1.381-28.494, p=0.017) and recurrence of disease (HR: 5.038, 95% CI: 1.327-19.131, p=0.018) (Table 4). In addition, high pathological T (pT) stage (HR: 32.956, 95% CI: 7.749-140.170, p<0.001) and low C-index value (HR: 0.352, 95% CI: 0.132-0.939, p=0.045) were predictive factors influencing RFS.

Discussion

Nephron-sparing approaches are focused solely on the removal of renal mass to optimise renal function. In this way, chronic kidney and cardiovascular diseases that may develop in the post-operative period have been prevented, thereby increasing the OS (10,11). The clinical and oncological effects of PSM after PN have been addressed. In recent years, studies which showed that PSM does not have oncologically worse effects have increased. Our aim was to reveal the clinical and oncological effects of PSM in our clinic with long follow-up data.

Previously, a healthy tissue margin of around 1 cm was proposed to get rid of cancer (12,13). However, some reports have indicated that the width of this healthy tissue margin does not affect oncological outcomes (13). Moreover, broader excision limits have been correlated with increased loss of normal parenchymal tissue volume and reduced renal function (14). In addition, some reports showed that simple enucleation and enucleo-resection techniques have equivalent oncological results compared to standard resection (15-17). Therefore, the absence of microscopic malignant tissue at the resection margin is known as NSM.

The occurrence of PSM reported in the literature ranges from 0% to 15%, regardless of the PN technique (18). The rate in our study (12.6%) did not vary from these series. Shah et al. (4) stated that PSM was irrelevant to tumour size, histology, localisation, nuclear grade, tumour stage or laterality. Conversely, another study identified blood loss, tumour grade and tumour stage as predictors of PSM (19). Ani et al. (20) indicated that there was an important correlation between stage and fat invasion and PSM. In the present analysis, we determined that the RENAL nephrometry score and C-index value were significant predictive factors for PSM. Most of literature studies report that PSM does not increase recurrence (21,22). Similarly, we also found that the prevalence of recurrence in PSM patients was higher, but not statistically significant. Some studies have reported that PSM is a poor prognostic factor for recurrence (3,7,23). Khalifeh et al. (3) also stated that they did not find a significant risk factor for PSM in the analysis of factors such as tumour size, grade, stage and surgeon’s learning curve. However, they concluded that PSM is associated with local recurrence and metastasis and reported a 3-year cancer-free recurrence rate of 47.0% and a 3-year metastasis-free survival rate of 63.0% in patients with PSM (3). Similarly, Bensalah et al. (6) reported that PSM had a higher risk of recurrence and a lower RFS in a study of 775 patients, of which 111 were patients with PSM. No difference was observed between the two groups on OS, CSS and RFS in the current study.

The uncertainty about PSM remains, as most studies have heterogeneous masses, fewer patients or shorter follow-up periods. In this study, the C-index, RENAL nephrometry and Padua scores, which define the tumour complexity, were analysed to determine the sample homogeneity. Fuhrman nuclear grade, tumour stage according to TNM staging, necrosis and capsule invasion were not statistically different between the groups. The C-index value, which predicted that patients with PSM had more complex tumours, was lower in group 1. We also found that C-index value predicting RFS was an independent risk factor in multivariate model analysis between groups that were homogenous in terms of tumour characteristics. Another study showed that the high RENAL nephrometry score was associated with an increased risk of residual disease (4). In addition, it was also reported that the risk of relapse in PSM patients was correlated with an increased pT stage or Fuhrman grade (4,6). In the present analysis, the advanced pT stage had a worse effect on CSS and OS in univariate analyses. On the other hand, Yossepowitch et al. (5) stated that intraoperative tumor control would be more difficult due to a decrease in tumor size and it would have a higher PSM rate. Furthermore, in a retrospective review of 1048 open PN patients by Patard et al. (24), tumour size did not affect the incidence of PSM.

Some studies have documented that renal ischaemia caused by clamping of the renal artery might destroy cells with rapid metabolic cycles, such as cancer cells (5). We also observed similar findings. Non-ischaemic procedures were mostly performed in patients with PSM in the current cohort. In addition to tumour foci that are extirpated with renal ischaemia, we thought that a decrease in visual quality due to bleeding in non-ischaemic procedures may also have an impact on this issue. Additionally, residual tumour cells have also been documented to have been damaged by thermal effects, such as cauterisation after renal mass resection (5). The residual tumour rate was stated to be 7%-39% in patients with PSM (6,25). In fact, this high incidence may also cause concern about the prognosis of patients. However, as in the current study, PSM was not associated with poor prognosis (5,21,22). Certain studies have shown a higher prevalence of high-grade RCC in patients with PSM (4,6,25). They also argue that a reduction in recurrence-free and metastasis-free survival in patients with PSM can be prevented through safer and wider resection, particularly in high-risk patients.

The treatment of PSM patients is also uncertain. It is controversial whether follow-up, total nephrectomy or re-resection should be done in patients with recurrent tumours. Indeed, Sundaram et al. (25) reported a study involving 29 PSM patients. No tumour was detected in any of the 8 patients who underwent complementary total nephrectomy; only two of the 21 patients who underwent re-resection were confirmed to have tumours (25). Retrospectively, Raz et al. (26) analysed 114 patients who underwent PN, 15% of whom had PSM. Approximately half of the patients with PSM underwent radical nephrectomy. Only 11.7% of these patients had residual tumour tissue. Complementary surgery for patients with PSM showed that it was overtreatment (26). Similarly, Yossepowitch et al. (5) reported that PSM was not a negative factor for local or metastatic progression in a cohort of 1344 patients for at 5 and 10 years of follow-up. In another study with a long follow-up period, it was confirmed that PSM did not pose a risk for local recurrence or distant metastasis (27). Moreover, active surveillance was reported to be more preferred than complementary nephrectomy or resection in the management of PSM patients (28). In our study, which also included long-term follow-up results, we found that PSM had no influence on OS, CSS or RFS. Reasonable options such as complementary radical nephrectomy, recurrent PN, energy ablation of the tumour bed or active monitoring for each patient should be applied to evaluate the patient and tumour characteristics (29).

We investigated the significance of PSM in RFS, CSS and OS analyses. Various literature studies have compared OS times of patients with PSM. Maurice et al. (30) determined that PSM had a significant hazard ratio of 1.35 for overall mortality. They also stated that PSM patients had older and higher CCI in univariate analyses and concluded that CCI and pT stage were associated with both PSM and OS in multivariate analyses. In comparison to our study, they emphasised that PSM was associated with poorer OS. In the multivariate analysis of the factors affecting the OS, we found that CCI and high pT stage were independent predictors. On the other hand, PSM was not effective on OS. In addition, Marszalek et al. (28) stated that PSM had no effect on disease-free survival and OS. Similarly, Bensalah et al. (6) found that PSM patients had no detrimental effect on progression-free survival and CSS. Moreover, there was a shorter time to recurrence in these patients than in NSM patients (21.3 vs 27.7 months, respectively; p=0.004), but there were no statistically significant differences in CSS. In the current study, we found that the independent factors influencing CSS were mid-renal localisation, high-stage tumour and recurrence.

Study Limitations

In addition to the retrospective design, this study had some limitations. There was no standard for tumour resection technique among surgeons as technological advances continued during the study period. Although the groups were identical in terms of tumour characteristics and demographic features, the absence of a complete matched pair was another limitation of our study. Propensity score matching helps to reduce selection bias and confounding. However, this study, which has longer follow-up times relative to many studies in the literature, contains an overview of data collected from the follow-up of the mid- to long-term oncological outcomes of PSM patients. In addition, one of the strengths of our study is that all cases belong to a single centre.

Conclusion

Although RCCs have heterogeneous characteristics, the probability of recurrence after PN is very low, as in groups with homogeneous tumour characteristics in our study. High RENAL nephrometry score and low C-index value suggest an increased risk of tumour complexity as an independent risk factor for predicting PSM. The C-index is also a predictor that affects RFS. Tumour stage is an independent risk factor that decreases both CSS and OS. Other predictive factors include CCI for OS, mid-renal localisation and recurrence for CSS and pT stage for RFS. Based on our findings, PSM is not a factor that has a detrimental effect on recurrence or survival, even though it is not a good pathology result after surgery. We therefore agree that an active and careful clinical monitoring could be the most effective method for patients with PSM.

Ethics

Ethics Committee Approval: Following the approval by the local ethics committee (approval date: 20.01.2020, decision number: 80/08).

Informed Consent: Retrospective study.

Peer-review: Externally and internally peer-reviewed.

Authorship Contributions

Concept: M.Ç.Ç., N.K., S.S., L.S., Design: M.Ç.Ç., N.K., S.S., H.T., H.E., Data Collection or Processing: M.Ç.Ç., N.K., A.K., S.S., F.S., O.R.K., H.E., Analysis or Interpretation: M.Ç.Ç., N.K., S.S., O.R.K., L.S., H.T., Literature Search: M.Ç.Ç., A.K., F.S., O.R.K., L.S., H.T., H.E., Writing: M.Ç.Ç., N.K., A.K., F.S., L.S., H.T.

Conflict of Interest: No conflict of interest was declared by the authors.

Financial Disclosure: The authors declared that this study received no financial support.