Is extending surveillance guidelines for renal cell carcinoma without understanding patient outcomes putting the cart before the horse?

Abstract

Treatment and surveillance guidelines for renal cell carcinoma (RCC) remain challenging, with a lack of conformity across oncologic and urologic societies. While debate continues regarding the benefits of active surveillance, partial or radical nephrectomy, or thermal ablation for small renal masses, a parallel discussion has emerged regarding the appropriate follow-up of RCC following definitive therapy. The challenges associated with post-treatment follow-up in oncology are not unique; best practices for follow-up after definitive therapy often lack level 1 evidence for many cancer sites. This absence of evidence leads to nonconformity among providers, confusion for patients, and overall concern that guidelines may be too stringent (or not stringent enough). The article that accompanies this editorial highlights the fact that strict attention to current guidelines misses a significant proportion of RCC recurrences. The authors examined post-treatment recurrences in 3,651 patients with RCC from a tertiary center, with minimal ( 3%) loss to follow-up and with standardized surveillance including quarterly imaging for the first 2 years, semiannually for the next 3 years, and annual surveillance thereafter. Surprisingly, nearly one third of recurrences were missed when following American Urological Association or National Comprehensive Cancer Network guidelines, with a need for significantly longer follow-up (up to 21 years) to capture 95% of these recurrences. Although these findings are striking and call into question whether modifications to existing surveillance guidelines should be made, several critical factors must be considered that we believe cast a different light on the potential impact of these results. First, findings from a single tertiary center may not be generalizable. The majority of nephrectomies are performed in the community, and populationbased treatment trends reveal that fewer partial nephrectomies are performed in this setting. The trend toward more partial nephrectomies in academic centers may explain why patients with low-risk disease undergoing radical nephrectomy have higher recurrence rates compared with those undergoing partial nephrectomy. As the authors hypothesize, perhaps tumors selected for radical nephrectomy in a specialized tertiary care center have features (eg, size, tumor location) that translate to worse biology (and higher recurrence rates). This effect may be muted in community centers, in which urologists are more likely to perform radical nephrectomy among patients with low (or high) -risk disease. Regardless of whether this hypothesis is true, higher recurrence rates among low-risk radical nephrectomy patients point to a larger issue: how accurate is our current characterization of lowand highrisk RCCs? Currently, surveillance guidelines use relatively crude estimations of recurrence risk, with tumor stage determining risk categories. Although tumor stage most certainly has value, the recent revolution in genomics has provided a new understanding of RCC biology that will undoubtedly improve risk stratification and subsequent design of surveillance guidelines across cancer types. A recent retrospective analysis evaluating the effect of BAP1 and PBRM1 mutations on survival in sporadic clear cell RCC found striking differences in clinical outcomes between these mutation-defined subtypes. Median overall survival for patients with BAP1-mutant tumors was significantly shorter when compared with those with PBRM1-mutant tumors, establishing a basis for molecular genetic classification as it relates to treatment and possibly surveillance decisions. Incorporating recently defined genetic signatures into existing guidelines is a logical next step. Validation of a 16-gene signature (Oncotype DX recurrence score) for prediction of recurrence after nephrectomy for stages I to III clear cell RCC was recently presented as a useful tool for surveillance strategies after surgery. In multivariable analyses, the gene signature remained a significant predictor for recurrence, even after adjustment for tumor size, Fuhrman grade, and necrosis. Incorporating molecular subtyping with prospective studies that evaluate recurrence and survival following definitive therapy for RCC will be needed to advance the field. Moving beyond a better understanding of baseline tumor biology, we also need to understand biologic differences in early and late recurrences and how they affect survival. If a shorter time to recurrence predicts for a reduced cancer-specific survival, guidelines may need to be designed to reflect the more indolent natural history of a late recurrence. Follow-up beyond a specified time period may not be necessary, even if some recurrences are missed. This highlights the most important issue of how surveillance affects our patients, both in terms of survival as well as quality of life and cost. Unfortunately, we have a poor understanding of whether surveillance following definitive therapy for RCC translates to a survival JOURNAL OF CLINICAL ONCOLOGY E D I T O R I A L VOLUME 32 NUMBER 36 DECEMBER 2