Adjuvant chemotherapy selection for stage I NSCLC 2011 through 2014 in the southeastern United States.

Abstract

248 Background: Among patients with early stage non-small cell lung cancer (NSCLC), appropriate use of adjuvant chemotherapy is an important determinant of survival. Existing level 1 data, which includes analysis of five cisplatin-based phase III trials (IALT, BLT, ALPI, ANITA, NCIC-CTG-JBR.10), two carboplatin-based phase III trials (CALGB 9633 and NATCH Trial), and two meta-analyses (LACE, MRC), demonstrate only cisplatin-based adjuvant chemotherapy regimens enhance overall patient survival. Most importantly, these studies also reveal that survival advantage from adjuvant chemotherapy does not extend to stage I patients, but only to stage II-IIIA patient subgroups. Unplanned a posteriori subset analyses indicate the possibility that certain stage I NSCLC with one or more "high risk" characteristics may benefit from cisplatin-based adjuvant chemotherapy. This hypothesis must be properly tested before it can be adopted. We, herein, hypothesize that many US stage I NSCLC patients are routinely treated in the adjuvant setting with regimens that have not demonstrated level one overall survival benefit. Methods: We examined chemotherapy request data for stage I & II NSCLC, submitted to a private health insurer by oncologists for the adjuvant treatment of NSCLC patients in the southeast US from January 2011 through May 2014. Results: Overall, 25% (67/264) of adjuvant chemotherapy requests were for stage I NSCLC patients (potential overuse). The frequency of these requests has increased over time: 17% in 2011, 20% in 2012, 30% in 2013, 34% in 2014. Four out of five of patients with stage I NSCLC (52/67) were given carboplatin-based regimens, while only one in five (19%, 13/67) received cisplatin-based regimens, and 3% (3/67) were pemetrexed-based regimens (potential misuse). Conclusions: High misuse and overuse rates of adjuvant chemotherapy for stage I NSCLC are observed among the privately insured populations of Florida, Texas and Georgia. Quality improvement initiatives in this patient population may be indicated to prevent potential harm from treatments advised against by ASCO, ESMO, NCIC, NCCN and UK guidelines.