SUO 2021: The Future of Adjuvant Therapy for High-Risk Non-Metastatic Renal Cell Carcinoma

(UroToday.com) The Society of Urologic Oncology (SUO) 2021 annual meeting in Orlando, FL hosted a thought-provoking presentation by Dr. Pavlos Msaouel, MD, Ph.D. regarding the future of adjuvant therapy for high-risk non-metastatic renal cell carcinoma (RCC). Dr. Msaouel began his talk by presenting the seminal data from KEYNOTE-564 evaluating adjuvant pembrolizumab (KEYTRUDA ®) after nephrectomy in RCC.

In this trial, disease-free survival was significantly improved in the intervention arm (HR 0.68, 95% CI: 0.53-0.87, p=0.002). The overall survival was similar in favor of the pembrolizumab arm (HR 0.54, 95% CI:0.30-0.96), with 18 deaths in the intervention arm compared to 33 deaths in the placebo arm. With regards to toxicity, patients in the intervention arm were more likely to require steroids following randomization (7.4% versus 0.6%) and Grade 3 or higher adverse events of any cause occurred in 32.4% of the patients who received pembrolizumab and in 17.7% of those who received placebo¹.

Next, Dr. Msaouel highlighted the underlying concepts behind current methodologic challenges in efficacy outcomes. There exist both direct and indirect effects of treatment choice on patient survival. As depicted in the image below, the direct effect is self-explanatory. The indirect effect is affected by treatment choice combined with potential subsequent disease recurrence, and potential additional systemic therapies prior to the event of interest (typically mortality).

If disease-free survival (DFS) and overall survival (OS) point in the same direction → Great
If DFS is negative but OS is positive → Drug is likely inert
If DFS is positive but OS is negative → Drug is likely active

Based on existing literature, survivors with localized RCC equally value DFS and OS outcomes, which is critical for trial design. The open challenge thus becomes to debias OS.

A key point of emphasis is that observational data should drive baseline risk estimates (i.e. risk of an event occurring without any intervention), whereas interventional data should be used to measure treatment effect (as opposed to the retrospective, observational data). More effort should be made to improve our prognostic risk models using representative real-world observational data and improve recurrence risk estimations using genomics/omics, radiomics, histomorphology and potentially circulating tumor DNA and cell-free methylated DNA.

Currently available prognostic tools include the ASSURE nomogram which risk stratifies RCC patients based on clinicopathologic features. It was developed using contemporary prospective randomized controlled trial data but analyzed as observational. It is freely available online and can be used to estimate 2-year DFS with no therapy².

RCC treatment effect calculators are also available and are depicted in the image below. However, they are restricted to patients with clear cell histology, which makes estimating treatment effects in non-clear cell histologies a challenging prospect.

Dr. Msaouel went on to give four case examples that highlight the shortcomings in current predictive/treatment RCC models.

The first patient is a 48-year-old female with a 7 cm pT3a ccRCC, grade 2 with renal vein invasion (pT3aN0M0). This is a KEYNOTE-564 eligible patient at intermediate-high risk who has a 2-year PFS with surveillance of 87.2% and a 2-year DFS of 91.1% with adjuvant pembrolizumab. Although the patient is eligible for pembrolizumab, based on the modest risk improvement of only 3.9% at two years, Dr. Msaouel would not advocate for adjuvant therapy in this situation.
The second patient is a 55-year-old male with a 10.3 cm pT2b ccRCC, grade 3, with necrosis (pT2bN0M0). This patient would not have been eligible for the KEYNOTE-564 trial, however, she would have had a significant estimated treatment benefit with adjuvant pembrolizumab (68% to 76.9%), thus making her an attractive candidate for adjuvant treatment per Dr. Msaouel. The open challenge in this situation is to improve model discrimination.
The third patient is a 52-year-old female with a 10.2 cm pT3a ccRCC, grade 3 with renal vein invasion, and no evidence of residual metastatic disease after solitary right upper lobe metastasectomy (pT2bN0M1 NED). This patient would have been trial eligible, however, based on available data, 2-year DFS with or without adjuvant therapy is not estimable and thus no recommendations could be made. The open challenger here is to develop risk nomograms for M1 NED patients.
The fourth patient is a 54-year-old female with a 13.9 cm pT3a papillary type 1 RCC, grade 2 with renal vein invasion (pT3aN0M0). This patient would not have been trial-eligible due to her papillary histology, thus accurate estimates of treatment effects with pembrolizumab are not realistically feasible (an estimated improvement from 94.6% to 97.3% based on best outcome probabilities). The open challenge here is to estimate treatment effects for papillary type 1 RCC.

Dr. Msaouel concluded his talk with the following summary research needs:

Statistical research:
- Develop a methodology to debias OS outcome
- Needs to be efficient, general, easy to implement, and interpretable by both the statistical and medical communities
Interventional research:
- The estimated treatment effect for rare histologies
- New adjuvant regimens
Observational research:
- Improve recurrence risk estimations: genomics/omics, radiomics, histomorphology
- Develop contemporary risk nomograms for M1 NED scenarios
- Develop risk nomograms for rare histologies

Presented by: Pavlos Msaouel, MD, Ph.D., Assistant Professor, Department of Genitourinary Medical Oncology, Division of Cancer Medicine, The University of Texas MD Anderson Cancer Center, Houston, TX

Written by: Rashid Sayyid, MD, MSc – Urology Chief Resident, Augusta University/Medical College of Georgia, @rksayyid on Twitter during the 2021 Society of Urologic Oncology (SUO) Winter Annual Meeting, Orlando, FL, Wed, Dec 1 – Fri, Dec 3, 2021.

References

Login