ASCO 2017: Survival outcomes in elderly patients with muscle invasive bladder cancer: An analysis of the National Cancer Database
While patients with bladder cancer are typically in their sixties to seventies, the authors of this study utilized the recently popular National Cancer Database (NCDB) to specifically assess survival outcomes in the oldest patients with MIBC. Specifically, they use the NCDB to assess outcomes of chemoradiation therapy (CRT) vs. radical cystectomy (RC) in elderly patients, based on the presumption that RC is often precluded in the elderly patients with multiple comorbidities. Additionally, they assess subpopulations that received radiotherapy (RT) or chemotherapy (CT) alone.
The NCDB is a dataset that reportedly captures more than 70% of incident neoplasms in the United States, comprising more than 29 million unique cancer cases. However, as many studies using the dataset have now pointed out, it is limited by the fact that it only reports overall survival and no cancer-specific outcomes.
They included all patients with all newly diagnosed MIBC (cT2-T4a N0 M0) who were aged 80 years or older at the time of diagnosis from 2004 to 2013. All included patients underwent TURBT followed by RC, RT alone (> = 50Gy), CT alone, CRT or no further treatment. As mentioned above, due to the limitations of the database, overall survival was the primary outcome.
Based on their inclusion criteria, 10,055 patients were identified (1,588 RC, 839 RT, 1,013 CT, 1,035 CRT, 5580 no further treatment) and 82.1% of patients had Stage II disease. Mean follow-up was 27.6 months (range, 0-137.8). While mean survival of the entire cohort was 28.0 months, mean survival of the RC and CRT arms was not significantly different: 43.8 months for patients with RC, 40.8 months for CRT (p = 0.85).
Five-year OS for CT alone, RT alone, CRT, and RC was 15.7%, 15.1%, 24.0% and 29.6%, respectively. Multivariate analysis adjusting for age, year of diagnosis, gender, race, treatment location, treatment facility, insurance status, Charlson/Deyo Score, grade, and stage demonstrated an OS benefit in favor of CRT (HR 0.70; 95% CI, 0.63-0.77; P < 0.0001) and RC (HR 0.69; 95% CI, 0.63-0.76; P < 0.0001) compared to other treatment arms.
Limitations / Discussion Points:
1. As mentioned above, the NCDB only captures overall survival. Without cancer-specific survival, it is hard to make any significant statements regarding the effect of the disease process on the survival outcomes.
2. With a range of 0-137.8 months follow-up, it is unclear how many of the patients had sufficient follow-up for analysis. A sensitivity analysis for patients with at least 6-12 months follow-up would have been better.
3. Intent of treatment is hard to capture in these retrospective studies. Treatment with radiotherapy may have been palliative at the time of administration, rather than curative. As such, this would skew the results negatively towards RT.
4. With >50% of patients not receiving treatment, it is uncertain how reliable these findings are.
5. Selection bias likely played a significant role in treatment choice, and since OS (rather than CSS) is the primary outcome, the differences in cancer-specific outcomes are impossible to asses. Healthier patients underwent either CRT or RC, while unhealthier patients likely underwent RT, CT or no therapy.
Unfortunately, due to the significant flaws in the construction of this study, it is not clear what the results add to the discussion of CRT vs. RC in elderly patients. These results are unlikely to change management.
Presented By: Benjamin Walker Fischer-Valuck
Co-Authors: Yuan Rao, Soumon Rudra, Caressa Hui, Brian Baumann, Hiram Alberto Gay, Jeff M. Michalski
Institution(s): Washington University School of Medicine in St. Louis, Department of Radiation Oncology, St. Louis, MO; Washington University in St. Louis, St. Louis, MO; Saint Louis University School of Medicine, St. Louis, MO; Washington University School of Medicine in St. Louis, St. Louis, MO
Written By: Thenappan Chandrasekar, MD, Clinical Fellow, University of Toronto, Princess Margaret Cancer Centre
Twitter: @tchandra_uromd
at the 2017 ASCO Annual Meeting - June 2 - 6, 2017 – Chicago, Illinois, USA