ASCO GU 2018: Utilization and Survival Implications of a Delayed Approach to Targeted Therapy for Metastatic Renal Cell Carcinoma: A Nationwide Cancer Registry Study
Preliminary data suggests TT may be safely delayed in appropriately selected patient, thereby delayed the start of the clock on time to failure. However the utilization and impact of delayed TT has not been evaluated on larger-scale.
The authors approach this from a population-level analysis using the National Cancer Database (NCDB). While this database has been used frequently in the past few years, it is significantly limited by lack of cause-specific survival data – hence, all outcomes are based on overall survival. As can be imagined, in the elderly population, this can make conclusions difficult to draw.
Knowing this, the authors queried the NCDB for patients with mRCC treated with cytoreductive nephrectomy and TT from 2006-2012. Time to initiation of TT was defined as ‘early’ (within 2 months), ‘moderately delayed’ (2-4 months), and ‘delayed’ (4-6 months), and ‘late’ ( > 6 months) based on time from diagnosis to initiation of therapy. This obiously infers intent of the physicians, but due to lack of granularity, the intent will never be truly known.
Multivariable logistic regression was performed to determine factors associated with delayed TT. The impact of time to initiation of TT on OS was estimated by Kaplan-Meier and Cox multivariable survival analysis. Again, cancer-specific survival was not able to be assessed due to the limitations of the databse.
Of the 2,716 patients in the analysis, the median (interquartile range [IQR]) follow-up was 18.8 (9.1-32.9) months, and 71.8% of patients had died at last follow-up. The median (IQR) time from diagnosis to initiation of TT was 2.1 (1.3-3.23) months, with the longest delay being 20.1 months.
1,255 patients (46.2%) had early TT, 1,072 patients (39.5%) had moderately delayed TT, 284 patients (10.5%) had delayed TT, and 105 patients (3.9%) had late TT. Delay in TT was not found to be a predictor of mortality in multivariable analysis; early TT (reference), moderately delayed TT (HR 0.98, p = 0.74), delayed TT (HR 0.95, p = 0.51), and late TT (HR 0.86, p = 0.20).
Time from diagnosis to initiation of TT and time from initiation of TT to patient death were not correlated after control for covariates (r= 0.04, p = 0.08).
Based on this, the authors conclude that delaying initiation of TT for mRCC following cytoreductive nephrectomy was not an independent predictor of worse OS, albeit in a highly selected study with inherent retrospective bias. Although this study is subject to limitations of observation study design and selection bias, the results are consistent with the notion that in carefully selected patients, outcomes might not be compromised with initial observation. We agree that prospective, randomized evaluation is warranted.
Limitations / Discussion Points:
1. This excludes patients who never received nephrectomy. This represents a population of mRCC patients that were not assessed.
2. This also excludes patients who underwent mRCC but who never made it to targeted therapy.
3. Delays due to medical comorbidities, inability to undergo surgery early, etc are not captured in this study. Fails to truly capture cause-specific mortality in a competing risks manner.
Speaker: Solomon Woldu
Co-Authors: Justin T. Matulay, Timothy N Clinton, Nirmish Singla, Yuval N Freifeld, Oner N Sanli, Laura-Maria Krabbe, Ryan C Hutchinson, Yair Lotan, Hans J. Hammers, Raquibul Hannan, James Brugarolas, Aditya Bagrodia, Vitaly Margulis
Institution(s): UT Southwestern Medical Center, Dallas, TX; Columbia University Medical Center, New York, NY; Johns Hopkins University Sidney Kimmel Comprehensive Cancer Center, Baltimore, MD
Written by: Thenappan Chandrasekar, MD, Clinical Fellow, University of Toronto, Twitter: @tchandra_uromd at the 2018 American Society of Clinical Oncology Genitourinary (ASCO GU) Cancers Symposium, February 8-10, 2018 - San Francisco, CA