Post-Metastasectomy Adjuvant Therapy in Patients with Renal Cell Carcinoma: A Systematic Review

INTRODUCTION

Metastasectomy is a key treatment option for patients with oligometastatic renal cell carcinoma (RCC), with the most common sites of non-kidney resection being the lungs, bone, lymph nodes, adrenal glands, pancreas, and liver, respectively [1, 2]. The optimal therapy for patients with metastatic renal cell carcinoma rendered free of disease after resection, often called “M1 NED patients”, is unclear. NCCN guidelines recommend adjuvant pembrolizumab within 1 year of nephrectomy for those patients with metastatic clear cell RCC (ccRCC) following complete resection of disease [3].

In the last four decades, despite sustained effort in the adjuvant setting, only two positive trials have been reported to improve disease-free survival (DFS). Among targeted therapies, the tyrosine kinase inhibitor (TKI) sunitinib was shown to modestly improve DFS, and among immune checkpoint inhibitors (ICIs), pembrolizumab was shown to significantly improve DFS and overall survival (OS) as part of the Keynote 564 trial [4–7]. Other systemic therapy trials have had less favorable results, including those with tyrosine kinase inhibitors (sorafenib, pazopanib, axitinib), and mTOR inhibitor (everolimus) [8–12], as well as those with immune checkpoint inhibitors (atezolizumab, ipilimumab+nivolumab, and nivolumab) [13–15]. Differences in efficacy between these trials have led to speculation regarding the activity of these drugs, specifically in the context of immunotherapy. Even though both pembrolizumab and nivolumab are programmed death-1 (PD-1) inhibitors, the former has been shown to bind to the PD-1 receptor with higher affinity [16]. Further, drugs such as atezolizumab inhibit the programmed death-ligand 1 (PD-L1) receptor instead, and this drug is not used in the context of kidney cancer due to a lack of overall survival benefit in phase-III clinical trials [17]. Other factors include, differences in inclusion criteria (inclusion of non-clear histologies in the IMmotion010 (atezolizumab) trial, duration of adjuvant therapy (six months in the CheckMate-914 trial (ipilimumab+nivolumab) vs. one year in KEYNOTE-564 trial (pembrolizumab).

Among adjuvant-treated patients, those with completely resected metastatic disease – i.e., M1 NED patients - are at a uniquely higher risk for progression. Within Keynote 564, the M1 NED subgroup had the lowest DFS HR for pembrolizumab compared to placebo in a sub-group analysis, suggesting that this sub-group may benefit most from adjuvant therapy [4]. The NCCN recommendation in favor of adjuvant pembrolizumab is based on these results.

While several adjuvant studies, including S-TRAC, EVEREST, and Checkmate 914, did not include the M1NED high-risk subgroup [7, 12, 13]. Several others did. in this review, we sought to systematically provide a review of the existing trials of systemic therapy after complete resection of metastatic RCC.

METHODS

Search strategy

A systematic literature search was performed in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 guidelines [18]. A search was performed in electronic databases, including PubMed/MEDLINE and Scopus. The search terms included (Renal cell carcinoma OR kidney cancer) AND (adjuvant AND tyrosine kinase inhibitor OR sunitinib OR sorafenib OR pazopanib OR axitinib OR everolimus OR immune checkpoint OR nivolumab OR ipilimumab OR pembrolizumab OR avelumab OR atezolizumab OR IL2) AND (metastasectomy or M1NED). An additional manual search was performed to identify published abstracts from the European Society of Medical Oncology and the American Society of Clinical Oncology (ASCO) annual meetings. We excluded non-English articles, articles without original data (e.g. review articles or editorials), and repeated publications on the same cohort. In multiple publications on the same trial, we reported the most recent publication with the longest follow-up. Ultimately, 9 papers were included in the final review, as shown in Fig. 1.

Fig. 1

PRISMA flowchart.

RESULTS

DISCUSSION

Among resected RCC patients, those with completely resected metastatic RCC have demonstrated systemic disease and have a high risk of recurrence. Many adjuvant therapy trials have not included these M1 NED patients. The more pronounced DFS HR of the small M1 NED subgroup within Keynote 564 suggested this patient population had a more pronounced benefit from adjuvant immunotherapy. Despite the impressive results of pembrolizumab, other adjuvant immunotherapy trials, IMmotion 010 and PROSPER did not show similar results. While an argument can be made for these trials having smaller subgroups, the IMmotion 010 had a higher proportion of M1 NED patients (14% versus 6% in Keynote 564), and one would have expected to see similar results between the two studies. The discrepant results may reflect differences in the agents used, pembrolizumab acting on PD-1 vs atezolizumab acting on PDL-1, especially as atezolizumab in combination with bevacizumab has not demonstrated OS benefit in metastatic ccRCC [17]. This has previously been demonstrated in a pan-tumor meta-analysis where, anti–PD-1 drugs were shown to be clinically superior to anti-PD-L1 drugs and has been proven using molecular analyses [25]. PROSPER, another clinical trial that involved neoadjuvant priming with nivolumab followed by adjuvant nivolumab after nephrectomy, was negative, though it included only 28 total patients with M1NED, thus making it difficult to draw meaningful conclusions. Prior to the advent of immune-checkpoint inhibitors, other cytokine based therapies such as interleukin-2 (IL2), interferon (IFN)-α had been the standard of care based on improved survival in metastatic RCC [26]. However, subsequent evaluation in the adjuvant setting did. Not show clinical benefit and these drugs were not adopted for use after nephrectomy or complete metastasectomy [20].

The two trials that evaluated targeted therapies, specifically tyrosine kinase inhibitors [RESORT (sorafenib) and the ECOG-ACRIN trial E2810 (pazopanib)] were specifically designed for patients who had undergone metastasectomy. They were both negative trials, suggesting a limited role for TKIs in resected metastatic disease. Mechanistically, it maybe because immunotherapy, which has been shown to yield prolonged complete responses in a subset of patients with gross metastatic disease, may also be more likely to eradicate micrometastases as compared to TKIs.

In addition to the studies presented in this systematic review, there are ongoing trials that include M1NED patients. The pending LITESPARK-022 trial, comparing pembrolizumab plus belzutifan or pembrolizumab plus placebo, will provide unique insights into whether the addition of belzutifan to pembrolizumab will yield extra benefit in this patient population [27].

The major unanswered question of adjuvant therapy in RCC is identifying which patients are cured with surgery alone and which patients are at high risk for recurrence. Clinical models exist, but predictive biomarkers are not yet established. Tissue-based genomic and transcriptomic biomarkers that may be associated with the risk of recurrence are being investigated [28]. Additionally, circulating tumor DNA, which has shown promise in other cancers, is being investigated to detect minimal residual disease in RCC. Currently, however, RCC is considered a low-shedding tumor, and circulating tumor DNA is often only detected in those patients with larger burden of disease [29, 30]. Ongoing efforts to develop more sensitive assays should aim to circumvent this shortcoming.

Another important aspect of adjuvant treatment in resectable RCC is that it should maximize the proportion of patients cured without adding undue burden. It is important to consider that in the Keynote 564 trial, the rate of discontinuation of drug due to any adverse event was higher in the pembrolizumab arm (21.1% vs. 2.2%). Patients in the pembrolizumab arm experienced a higher incidence of serious adverse events of any cause (20.7% vs. 11.5%), as well as drug-related high-grade adverse events (18.6% vs. 1.2%) [5]. Thus inclusion of medical side effects of treatment as well as the financial toxicity is important to consider, as adjuvant pembrolizumab has only been found to be cost-effective in the highest-risk patients [31].

This systematic review has several limitations. As has been shown in previous studies, prognosis of patients with kidney cancer is dependent on additional clinical factors such as metastatic sites involved, and time since nephrectomy [32, 33]. However, granular details regarding sites of metastases resected, number of lesions subjected to metastasectomy, and interval since nephrectomy were not available in the papers cited from clinical trials in this review and should be a topic of a future analysis if such details could be made available. While searching literature for evidence, we found that some studies that did not report numerical data on survival and in the context of M1NED patients. Therefore, due to lack of data, a merged analysis of outcomes was not possible. Moreover, this being a systematic review, data was collected in a retrospective manner, and was thus prone to errors. There are also prior studies to suggest that there can be a publication bias associated with systematic reviews, in that prior studies with only significant results may get over-selected when gathering evidence [34].

Further studies to evaluate the role of adjuvant systemic therapy in M1NED patients should focus on the incorporation of biomarkers to identify patients at the highest risk of relapse, whether in the form of minimal residual disease assessment from circulating tumor DNA analysis or refined molecular techniques, such as transcriptomic signatures. Further, the inherent risk of subjecting these patients to adverse events from these systemic drugs should be considered when designing future trials for this patient population that may have been cured with metastasectomy alone.

CONCLUSION

No adjuvant TKI trial has shown improvement in DFS for M1 NED patients with ccRCC. The markedly lower DFS HR for the M1 NED subgroup for pembrolizumab within Keynote 564 has not been replicated in other immunotherapy trials using nivolumab and atezolizumab. The Chekmate-914 trial, that tested the combination of nivolumab plus ipilimumab and did not include M1NED patients, was speculated to be negative due to the lack of oligometastatic disease patients when compared to the pembrolizumab trial; the present review however points out that trials that had included a similar population were also negative. The conflicting results within adjuvant immunotherapy trials suggest more research is needed to determine which patients are at highest risk of relapse and would thus benefit most from adjuvant immunotherapy, specifically pembrolizumab.