A Multicentered,Propensity Matched Analysis Comparing Laparoscopic and Open Surgery for pT3a Renal Cell Carcinoma

Introduction

Laparoscopic renal surgery (LRS) has become a well-established tool in the surgical management of renal cell carcinoma (RCC). The reduced morbidity over conventional open renal surgery (ORS) techniques are well recognized with improved pain control, faster recovery, and improved surgical site cosmesis. ^{1 –4} Many studies have attempted to compare the oncologic efficacy between LRS and ORS, although these have primarily been focused on those with T1 or T2 disease. ^2,4 Improvements in surgical innovation and operator experience have led to expanding indications for LRS, yet, there is a paucity of literature evaluating oncologic efficacy to ORS in this population. ^{5 –7} Pathologic T3a (pT3a) disease includes renal tumors with segmental vein, renal vein, or perirenal adipose tissue invasion. This advanced disease state is typically aggressive with 5-year disease-free survival rates ranging from 30% to 85%, ^6,7 including a median time to recurrence ranging from 11 to 22 months. ^{8 –11} Provided oncologic outcomes can be preserved, a minimally invasive approach would be preferred to reduce perioperative morbidity.

Some studies have examined the role of LRS in the pT3 RCC population, but these studies are limited to noncomparative, descriptive analyses from small, single-centered cohorts. ^{5 –7} Comparative studies have been limited to retrospective reviews, ^{12 –15} with only one study utilizing a prospective matched analysis of 25 pairs. ¹⁶ While encouraging, their small numbers limit the utility and generalizability of their findings. Furthermore, a comparative analysis allows for evaluation between LRS and ORS, adjusting for a number of variables permitting the closest matched comparison. ¹⁷ This is a useful statistical approach to examine differences in outcomes between the groups outside of a randomized controlled study. Using a propensity-matched approach, we sought to compare oncologic outcomes between LRS and ORS in the treatment of pT3a RCC, using a large, multicentered collaborative effort.

Materials and Methods

The Canadian Kidney Cancer information system (CKCis) database is a multi-institutional collaborative database that captures clinical, pathologic, and oncologic data from patients diagnosed and treated for RCC. This access-restricted database currently consists of data from 15 institutions in six Canadian provinces. The database includes baseline retrospective information with prospective data collection since January 2011. Individual patients provided consent for their inclusion in the database, and institutional review board approval was obtained from each contributing site. Patients who underwent either LRS or ORS for pT3a RCC between 2008 and 2015 were identified. LRS included both laparoscopic radical nephrectomy (LRN) and laparoscopic partial nephrectomy (LPN). ORS included both open radical nephrectomy (ORN) and open partial nephrectomy (OPN). The choice of procedure was not randomized and dependent on the preferences of the patient and surgeon. These factors generally include the following: history of abdominal surgery, tumor complexity, and medical comorbidities.

All patients underwent preoperative chest imaging and routine blood work, including complete blood count, extended electrolytes, blood urea nitrogen, serum creatinine, and liver function tests. Patients were excluded if they had metastatic disease at the time of surgery. Follow-up protocol was determined by the operating surgeon and was generally in keeping with the 2008 Canadian Urological Association guidelines on the “Follow-up guidelines after radical or partial nephrectomy for localized and locally advanced renal cell carcinoma.” ¹⁸ The database is prospectively updated every 6 months.

All patients were staged according to the American Joint Committee of Cancer Staging manual, seventh edition. ¹⁹ Propensity matching was chosen as it permits the comparison of outcomes in patients who are similar on all measured baseline characteristics, except for the exposure. As such, it is well suited to minimize bias in observational cohorts. Nearest neighbor without replacement propensity-score matching was performed for age, gender, year of surgery, tumor size, clinical T stage, grade, histology, and partial vs radical nephrectomy to compare treatment modalities (LRS vs ORS) resulting in balanced groups. The primary endpoint was recurrence-free survival (RFS), which we defined as the time from surgery to the development of metastatic disease, local recurrence, or cancer related death. Patients were censored at last follow-up or at the time of noncancer-related death. Log-rank testing was used to assess for differences in survival. The Kaplan–Meier method was used to estimate survival curves. Statistical significance was defined as p < 0.05, and analyses were performed using R statistical environment.

Results

During the study period, 3638 patients were surgically treated for RCC within the CKCis database. Of these patients, 501 were found to have pT3a disease and were free of nodal or metastatic disease at the time of surgery, of whom 226 (45.1%) underwent LRS, and 275 (54.9%) underwent ORS. Unbalanced covariate comparison is listed in Table 1. We then balanced the two groups based on our predetermined covariates, resulting in a cohort of 452 patients divided into two groups to facilitate our matched analysis. The two groups were balanced with respect to year of surgery, age, gender, serum creatinine, tumor size, surgical approach (partial or radical nephrectomy), clinical stage, tumor grade, and histology (Table 2).

Table 2 demonstrates the clinical and pathologic characteristics between treatment groups. The average age was 63, with the majority of patients being Caucasian males. The groups were similar in their preoperative renal function and clinical stage. Mean tumor size was 7 cm in both groups. The majority of patients were treated with radical nephrectomy (88.5% of the LRS group and 87.6% of the ORS group). Operative time was similar between both groups, although there was significantly less blood loss in the LRS group (100 mL vs 400 mL in the ORS group, p < 0.01). In terms of pathologic outcomes, overall positive surgical margin (PSM) was 7% and 13% in the LRS and ORS, respectively, p = 0.09. The majority of tumors were high grade, clear cell RCC.

After a median follow-up of 21.1 months, the overall RFS was not significantly different between the LRS and ORS groups (Fig. 1). Oncologic outcomes are described in Table 3. The 3-year RFS was 63% and 50% for LRS and ORS, respectively. Median time to recurrence was 4.1 years in the LRS and was not reached for the ORS group. When limited to those who underwent radical nephrectomy, the 3-year RFS for LRN was 61% compared with 46% in the ORN group (p = 0.32) with a median time to progression of 3.9 years following LRS and 2.1 years after ORS. Similarly, when limiting to patients treated by partial nephrectomy, the 3-year RFS was 77% in the LRN group and 79% in the OPN group (p = 0.82). No recurrences such as port-site metastasis or peritoneal/retroperitoneal carcinomatosis were observed during the study period.

OPEN IN VIEWER

FIG. 1.

Kaplan–Meier recurrence-free survival estimates for laparoscopic and open renal surgery in the management of pT3a renal cell carcinoma, p = 0.36.

A sensitivity analysis of patients with cT3 was performed to evaluate the RFS for the ORN (n = 59) vs LRN (n = 59) cohorts that demonstrated no statistically significant difference, which correlates with our overall findings.

Discussion

Minimally invasive surgery has gone through many years of advances and gained experience, particularly in the management of RCC. With growing experience, more advanced tumors are likely to be managed by this approach; however; there is limited data examining oncologic outcomes in the locally advanced RCC population. Our findings represent the largest comparative evaluation of patients undergoing LRS and ORS for pT3a disease in the literature. Using a large, multi-institutional design and propensity-matched scoring approach; we observed similar oncologic outcomes for laparoscopically treated patients compared with patients treated with open surgery.

The efficacy of laparoscopic surgery for locally advanced RCC has been described in smaller, noncomparative studies. Stewart and colleagues reviewed 77 patients who underwent laparoscopic management of pT3 RCC (36% pT3a and 62% pT3b) with curative intent. In their cohort, 28.6% of patients developed disease recurrence after a median follow-up of 13.9 months, resulting in an estimated 5-year progression-free survival rate of 44.6%. ⁶ Bansal and colleagues also published a study of 32 patients with localized pT3a and pT3b RCC treated by minimally invasive surgery. After a mean follow-up of 42 months, 56.2% of patients were free of disease. ⁷ Furthermore, we have also previously described our laparoscopic experience for 176 patients with pT3 disease and found a 3-year RFS of 67% and a median time to progression of 10.3 months. ²⁰ These noncomparative findings suggest that laparoscopic surgery may be a feasible treatment option for patients with locally advanced RCC.

In the present study, we confirm the findings from single center, retrospective series using a propensity matched approach as a comparative effectiveness tool that is best suited to adjust for selection biases within an observational cohort. Outside of a randomized controlled trial, propensity matching offers the ability to balance cohorts for known and measured confounders to reduce biases. Notably, this approach is limited by the inherent ability to only adjust for measured confounders, whereas a randomized clinical trial would also account for unmeasured confounders. Laird et al. compared 25 patients with pT3 disease to 25 open cases with their analysis matched for gender, histology, Tumor, node and metastasis (TNM) stage, Fuhrman grade, <3 cm difference in tumor size, and <10-year difference in age at time of surgery. At a median follow-up of almost 5 years, they found no difference in overall survival (LRN: 56.3%, ORN: 61.8%, p = 0.36), cancer-specific survival (LRN: 64.4%, ORN: 67.3%, p = 0.70), or progression-free survival (LRN: 60.0%, ORN: 67.6%, p = 0.97). ¹⁶ In addition, Bensalah and colleagues evaluated 44 cases of LRN compared with 135 cases of ORN for pT3 disease with groups similar for tumor size, T3 subclassification, grade, and histologic subtype. After a median follow-up of 28 and 55 months for the LRN and ORN groups, respectively, they found no difference in overall survival. ¹⁵ Thus, our findings add to the growing literature that oncologic control can be preserved using a minimally invasive approach to surgery.

We included a combination of radical and partial nephrectomy patients in our analysis with sensitivity analyses for these subgroups confirming our overall findings. Current guideline recommendations for clinically localized T3 disease support radical nephrectomy. When limiting our analyses to patients treated by radical nephrectomy, the 3-year RFS rates were slightly improved for our laparoscopic group, but this difference was not significant (3-years RFS 61% LRS vs 46% ORN, p = 0.32). This difference in RFS is likely related to unmeasured surgical selection confounders, including tumor location or complexity as well as differences in patient comorbidity profiles.

With respect to nephron-sparring surgery, the 3-year RFS was 77% and 79% for the LRS and ORS groups, respectively. The more favorable, yet statistically insignificant, RFS compared with the radical nephrectomy cohorts is likely due to preoperative surgical selection. For example, the proportion of clinical T1 tumors was much higher in the partial nephrectomy groups, which would indicate that these patients experienced pathologic upstaging. The rate of pathologic upstaging in our CKCis has been previously reported as 9%. ²¹ This finding was also present in our open group, which consisted of 79% cT1 masses in the OPN group but only 28% in the ORN group. Mean tumor size among the LRS group was 4.27 cm for patients undergoing LPN and 7.46 cm for those undergoing LRN. As for the ORS group, mean tumor size was 3.88 cm among the OPN group and 7.41 cm for the ORN group. Clearly, patients who underwent partial nephrectomy were more likely to have a favorable clinical stage, but unfortunately had an adverse final surgical pathology. As for the more favorable outcomes among patients treated with partial nephrectomy compared with open surgery, this is likely due to the importance of tumor complexity and size. Previous reports have suggested that oncologic outcomes of T3a tumor <7 cm are similar to those of a T2 lesion ²² and may even be similar to a T1 tumor. ²³ Ultimately, when offering a nephron-sparing approach, the goal should be complete tumor removal in the form of a negative surgical margin. Although a topic of controversy, more recent reports have suggested an increased risk of tumor recurrence and metastasis with a PSM. ^24,25 Our positive margin rate among our LPN and OPN groups were found to be higher than the published literature, and this is likely due to complex tumor anatomy.

Although not statistically significant, there was a trend for a higher PSM in the ORS group (LRS = 7%, ORS = 13%, p = 0.09). As the operating surgeon determined the surgical approach, it may be assumed that tumors with greater complexity on cross-sectional imaging were more likely to be managed from an open surgical technique and are more prone to pathologic upstaging. As per Table 1, there was a greater proportion of cT3 tumors in the ORS group (LRS = 26%, ORS = 33%), and although this was balanced using the propensity-matched scoring, it did not take into account tumor complexity, which is an important limitation of our study.

An important consideration when comparing surgical approaches is the perioperative morbidity. We found that surgical operating time was similar between groups, but laparoscopy was associated with significantly less blood loss. Specific complication rates and other indices of morbidity, including length of stay and postoperative analgesia requirements, are other notable outcomes that we did not evaluate in the current study. However, these outcomes have been shown by others to be superior with laparoscopic surgery. ^4,26 Rather, discrepancies may arise in oncologic control, for which we did not find any significant difference indicating that laparoscopic surgery is safe for the treatment of pT3a disease.

There are a number of limitations of this study to highlight. Our multi-institutional design permits heterogeneity in surgical techniques and management practices. In addition, we acknowledge that pT3 RCC includes a heterogeneous population, including those with either venous or fat invasion, which may be associated with differential prognoses. ^23,27 With respect to statistical analysis, propensity-matched scoring only takes into account measured variables, and therefore, unmeasured confounders such as tumor complexity, patient factors such as comorbidities, and prior surgical interventions are not taken into account. Although propensity-matched scoring can answer important clinical questions, it is not a substitute for randomization. ²⁸ In addition, the median follow-up in our study was 21.1 months, so that further cases of local recurrence or metastatic disease will be anticipated with longer follow-up. With respect to pathology, there was no centralized pathologic reporting, and as much, there are concerns with respect to standardization. Finally, since classification was based on the final surgical procedure, we were unable to capture the proportion of patients who initially had laparoscopic surgery that were converted to open surgery. This may select for inferior outcomes with open surgery.

Conclusion

We have described the largest matched cohort study examining outcomes between LRS and ORS for the treatment of pT3a RCC. In our experience, laparoscopic surgery provides similar cancer control compared with ORS for pT3a disease. A minimally invasive approach should be considered the standard of care for locally advanced RCC whenever technically feasible.