Introduction:

Oral targeted agents such as abiraterone and enzalutamide are increasingly used in the management of advanced prostate cancer. However, the high out-of-pocket costs associated with these drugs can pose barriers to drug access and continued use. The 340B drug pricing program allows eligible hospitals to purchase outpatient medications at steep discounts, generating millions of dollars in savings per hospital. These savings provide hospitals with opportunities to invest in safety net programs (e.g., drug discounts) aimed at improving patient care, particularly for those who have socioeconomic disadvantages. The extent to which hospitals expand their safety net and improve patient care using 340B savings is unclear. We assessed the effect of hospital 340B participation on the care of men with advanced prostate cancer.

Methods:

A 20% sample of national Medicare claims was used to perform a retrospective cohort study of men with advanced prostate cancer from 2012 through 2019 who received their care at a hospital. The primary outcome was the initiation of an oral targeted agent. Secondary outcomes included monthly out-of-pocket costs for targeted agents and adherence to therapy. Multivariable regression models with clustering at the hospital-level were fit to adjust for differences in patient and market characteristics. We assessed the effect of hospital 340B participation and socioeconomic status—measured by social vulnerability index quintiles—on targeted agent access, monthly out-of-pocket costs, and adherence. We also assessed the effect of an interaction term between hospital 340B participation and social vulnerability on each outcome.

Results:

There were 2,401 and 1,142 men with advanced prostate cancer who received care at a 340B participating and non-participating hospital, respectively. Hospital 340B participation did not affect targeted agent use (OR 0.92, p=0.38), while social vulnerability did (Figure 1). Men residing in the third (23% vs 30%, p=0.010) and fifth (25% vs 30%, p=0.030) socially vulnerable quintiles were less likely to be started on a targeted agent, compared to the first quintile (least vulnerable). The interaction between hospital 340B participation and social vulnerability (Figure 2) did not affect targeted agent use (p>0.05). Furthermore, hospital 340B participation did not affect out-of-pocket costs (coefficient 0.030, p=0.69) or adherence (OR 0.99 p=0.50). While social vulnerability was associated with variations in out-of-pocket cost—with those residing in the third quintile experiencing the highest out-of-pocket costs ($855)—adherence was not affected. The interaction term did not influence out-of-pocket costs or adherence (p>0.05).

Conclusion:

Despite the 340B program’s intent to bolster the safety net of participating hospitals, it did not affect the care of men with advanced prostate cancer. In light of growing calls from policymakers to reform the 340B program, due to concerns regarding its effectiveness, focus could first shift towards providing hospitals with guidance on how to effectively use savings to benefit those most in need. One practical option would be for hospitals to use 340B savings to provide drug discounts to those who are most susceptible to high drug costs (i.e., socially vulnerable men who do not qualify for payment assistance programs).

Funding: T32CA180984; R01CA275993