Category: Professional Posters
Purpose: Adverse events (AEs) related to cancer therapy reduce patients’ quality of life and generate substantial healthcare costs. There is limited real-world evidence regarding the AE costs for patients with metastatic pancreatic cancer (m-PANC) who receive FDA-Approved/NCCN Category 1 treatments. We analyzed the costs of three of the most common AEs for patients receiving FDA-Approved/NCCN Category 1 treatments: neutropenia, anemia, and thrombocytopenia in the Medicare fee-for-service (FFS) population by chemotherapy regimen and line of therapy (LOT).
Methods: We identified patients with m-PANC using ICD-9/10 diagnosis codes in the 2013-2017 Medicare 100% Limited Data Set claims, which include all Medicare FFS Part A and B claims, except professional services, for 45 million beneficiaries. Patients in our study had multiple claims with a pancreatic cancer (PANC) diagnosis more than 30 days apart and one+ claim(s) with a secondary malignancy (metastasis) diagnosis on/after the first PANC diagnosis date. We defined the index date as the earliest metastasis diagnosis date. We excluded patients with pre-index non-PANC malignancies and those without six-month pre-index and three-month (or until death, if earlier) post-index Medicare FFS enrollment. LOTs were assigned based on the order of therapies used. LOTs ended when a new regimen began, 28 days after the last chemotherapy (if no new regimen), or upon death.
We identified AEs using ICD-9/10 diagnosis codes occurring during LOTs. We randomly sampled control patients in the same LOT and regimen without an AE and assigned them shadow AE dates. We calculated 30-day costs after the AE onset or shadow AE date. For LOTs and regimens with at least 80 patients (cases and controls), we estimated 30-day AE incremental costs using log-link generalized linear models and gamma-distributed errors. We predicted mean 30-day AE incremental costs relative to controls using recycled projections and bootstrapped 95% confidence intervals to determine statistical significance relative to zero.
Results: Anemia was the most common AE, ranging from 41% of patients receiving first line (1L) gemcitabine/nab-paclitaxel to 32% of patients receiving second line (2L) liposomal irinotecan. Mean 30-day anemia incremental costs were $3,924 for 1L gemcitabine/nab-paclitaxel, $3,080 for 1L gemcitabine monotherapy, $3,725 for 1L FOLFIRINOX, and $3,257 for 2L liposomal irinotecan, all of which were statistically significant.
Neutropenia was observed for 20% of patients receiving 1L gemcitabine/nab-paclitaxel, 16% of those receiving 1L gemcitabine monotherapy, 32% of those receiving 1L FOLFIRINOX, and 19% of patients receiving 2L liposomal irinotecan. Mean 30-day neutropenia incremental costs were $2,503 for 1L gemcitabine/nab-paclitaxel, $1,610 for 1L gemcitabine monotherapy, and $2,322 for 1L FOLFIRINOX, all of which were statistically significant. The mean 30-day neutropenia incremental costs for (2L) liposomal irinotecan were $1,284, which was not statistically significant, possibly because there were few cases (42).
The occurrence of thrombocytopenia ranged from 18% of patients receiving 1L FOLFIRINOX to 8% of those receiving 2L liposomal irinotecan. Mean 30-day thrombocytopenia incremental costs were $2,678 for 1L gemcitabine/nab-paclitaxel, $3,291 for 1L gemcitabine monotherapy, and $3,721 for 1L FOLFIRINOX, all of which were statistically significant. There were too few 2L liposomal irinotecan patients with thrombocytopenia to estimate incremental costs.
Conclusion: AEs impose substantial costs for Medicare FFS patients with m-PANC receiving FDA-Approved/NCCN Category 1 treatments. For 1L regimens, we observed statistically significant incremental costs associated with three of the most common m-PANC AEs: anemia (mean incremental cost: $3,080-$3,924), neutropenia (mean incremental cost: $1,610-$2,503), and thrombocytopenia (mean incremental cost: $2,678-$3,721). Among 2L liposomal irinotecan patients, only anemia incremental costs ($3,257) were statistically significant; neutropenia incremental costs were not statistically different from zero, and there were insufficient thrombocytopenia cases to estimate incremental costs.