Abstract
Background: The introduction of novel parenteral (bortezomib) and oral (lenalidomide and thalidomide) agents in the mid-2000's radically improved the management and outcomes in multiple myeloma. The cost of these drugs typically exceeds $5000/month, posing a financial challenge for patients (pts), insurers and healthcare systems. Medicare beneficiaries in the United States have disparate coverage for parenteral and oral chemotherapy. Coverage for oral agents has been available since 2006 for those purchasing Part D prescription plans (PDP), but such plans impose substantial out-of-pocket expenses through the "donut hole" coverage gap and coinsurance required even in the "catastrophic" phase. Pts with low income can receive the Part D Low Income Subsidy (LIS) which largely eliminates this cost sharing. Our objective was to examine the association of Medicare part D and LIS policies with utilization of, and out-of-pocket burden for, the novel anti-myeloma agents.
Methods: From the Surveillance, Epidemiology and End Results data base linked to Medicare claims (2000-2012), we selected pts ≥65 years old, diagnosed with myeloma in 2001-2011, and with fee-for-service Medicare insurance. Pts in the post-Part-D era (2006-2012) were classified according to their prescription coverage at diagnosis as those with: PDP without LIS, PDP with LIS, or no coverage. We identified first-line parenteral chemotherapy (using specific drug codes) and oral agents (assumed from generic follow-up codes for pts without PDP), as well as other relevant covariates: comorbidities, a claims-based performance status indicator, anemia, neuropathy, and use of health services. We then compared parenteral and oral chemotherapy use between pre-Part-D (2001-2005) and post-Part-D (2006-2011) eras, and use of lenalidomide/thalidomide among PDP pts with or without LIS, in multivariable robust Poisson models for relative risk (RR), reporting 95% confidence intervals (CI).
Results: Among 19,434 pts (median age, 77 years; 46% women), 52% had a record of any chemotherapy within 1 year from myeloma diagnosis. This proportion increased from 42% in pre-Part-D era to 61% after Part D introduction (adjusted RR, 1.45; CI, 1.41-1.49; P<.0001), and bortezomib replaced other parenteral regimens. In the post-Part-D era, the proportion of pts receiving parenteral chemotherapy increased for those without prescription coverage (Fig. A, adjusted RR compared with pre-Part-D era, 1.30; CI, 1.23-1.38; P<.001), while it decreased, in favor of oral regimens, for those who acquired a PDP without LIS (RR, 0.90; CI, 0.85-0.95; P<.001), or PDP with LIS (RR, 0.85; CI, 0.78-0.92; P<.001). Among pts with no PDP coverage, 41% purchased it within 1 year from myeloma diagnosis.
We then directly analyzed the upfront use of novel oral agents among treated pts with a PDP (and thus complete record of prescriptions, N=3,142). Lenalidomide largely replaced thalidomide between 2007 and 2011, but up to 40% of pts did not receive novel agents as part of their regimen (Fig. B). Pts with LIS had the same median age as those without LIS, but were more often female, non-white, with poor performance, or comorbidities (all P<.0001). Adjusting for clinical differences, recipients of LIS had a 17% higher probability of being treated with a novel oral agent (RR, 1.17; CI, 1.05-1.29; P=.0034). Median patient cost-sharing for the first prescription was $3 for pts with LIS and $3,120 without LIS. At the start of oral chemotherapy, 3% of pts were in the catastrophic phase of PDP coverage, while 79% reached it with the 2nd prescription. Median number of prescription fills was 6 (range, 1-27), and median gross cost of lenalidomide/thalidomide within the 1st year of therapy was $43,920.
Conclusions: The proportion of Medicare beneficiaries actively treated with chemotherapy for myeloma increased after the introduction of Medicare Part D in 2006. Accounting for this, the availability of prescription coverage was associated with relatively increased use of oral regimens, and the use of novel agents was higher for those receiving Part D LIS which alleviates the out-of-pocket expenses. Our findings suggest concerning disparities in equitable access to novel, but expensive oral anti-myeloma agents based on insurance coverage for oral chemotherapy. Further research should determine if policies for oral chemotherapy coverage translate into disparities in survival or other outcomes.
Olszewski:TG Therapeutics: Research Funding; Genentech: Research Funding; Bristol-Myers Squibb: Consultancy.
