United States treatment pattern response to carboplatin and cisplatin shortages.

Abstract

11088 Background: In response to cisplatin (2/10/23) and carboplatin (4/28/23) shortages, national guidelines recommended prioritizing cisplatin and carboplatin for their most important indications to preserve availability while maximizing survival outcomes. We examined the real-world response to recent platinum shortages in lung cancers (LC), for which guidelines recommended preserving platinum chemotherapies when non-metastatic versus allowing substitution to immunotherapy monotherapy when metastatic. Methods: Using IQVIA PharMetrics data (9/1/22-10/31/23), we developed claims algorithms to identify intravenous anticancer therapy infusion claims for patients with incident non-metastatic vs. metastatic LC, which we validated against SEER-Medicare. Our outcomes were the fraction of infusions containing either (1) cisplatin, (2) carboplatin, or (3) immunotherapy (IO) without platinum. We performed an interrupted time series (ITS) analysis, using a linear probability model with three time-periods: pre-platinum shortage (9/1/22-1/31/23), post-cisplatin shortage (2/1/23-4/30/23), and post-carboplatin shortage (5/1/23-10/31/23), adjusting for age, gender, region, and comorbidities. Results: There were 11,983 infusions (2,411 patients) and 1,650 infusions (428 patients) for incident non-metastatic and metastatic LC, respectively. For non-metastatic LC, between 2/2023 (cisplatin shortage) and 04/2023 (carboplatin shortage), cisplatin use declined by 1.7 percentage point (pp) per month (p=0.03) and a non-significant 1.6 pp monthly increase in carboplatin usage (p=0.24). After carboplatin shortage, there was an immediate 4.5 pp drop in use of carboplatin (p=0.04), as well as an immediate 5.8 pp increase in use of cisplatin (p<0.0001). Overall platinum use (receipt of either cisplatin or carboplatin) throughout this entire period did not have any significant change. For metastatic LC, cisplatin was rarely used, so results after the cisplatin shortage were not assessed. After carboplatin shortage, there was a 15.7 pp immediate decline in use of carboplatin (p=0.002), with a concomitant 18.4 pp immediate increase in use of IO without platinum (p<0.001). Conclusions: The real-world response to platinum shortages in LC appeared to follow guidelines. For non-metastatic LC, for which platinum chemotherapy increases cure rates, declines during either the cisplatin or carboplatin shortages were equally balanced with increases in the other platinum chemotherapy, such that overall platinum chemotherapy usage remain stable. For metastatic LC, declines in carboplatin were balanced with increases in immunotherapy monotherapy.