Impact of Medicare Advantage (MA) on timely initiation of pembrolizumab among dMMR/MSI-h metastatic colorectal cancer patients.

• DOI: 10.1200/jco.2025.43.16_suppl.3548
• Publication Date: 2025-05-28T18:32:01Z
• Authors: Baqir Jafry; Chuan Angel Lu; Timothy J. Brown; Syed Mohammad Ali...; Changchuan Jiang

3548

Background: With MA plans covering over half of Medicare beneficiaries, concerns remain about their ability to manage complex cancer care due to pre-authorization and limited provider network. This study evaluates MA versus Traditional Medicare (TM) regarding timely initiation of Pembrolizumab for dMMR/MSI-H colorectal cancer (CRC), following its 2020 FDA approval. Methods: This study utilized nationwide Flatiron Health Electronic Health Record-derived de-identified database. We included patients diagnosed with dMMR/MSI-H CRC from 2020 onward, aged ≥65 years, who had at least one clinic visit within six months of diagnosis and were insured under MA or TM. The primary endpoint was initiation of Pembrolizumab within 30-, 45-, and 90-days post-diagnosis. Multivariable logistic regression with Inverse Probability Weighting, adjusted for SES (defined by Yost score), age, race, ECOG, practice type and diagnosis year, evaluated impact of insurance type on timely treatment initiation. Results: Out of 597 dMMR/MSI-H metastatic CRC patients identified since 2020, 219 had at least one clinical visit under MA (N = 86) or TM (N = 133) plans. Predominantly, patients in our cohort were non-Hispanic White (72%), with higher SES (59%), diagnosed with de novo dMMR/MSI-H (59%), and treated in community hospitals (86%). Of 95 patients who commenced pembrolizumab as initial therapy, 62 (65%) had TM and 33 (35%) had MA. The adjusted analysis revealed that MA patients were significantly less likely to start pembrolizumab within 90 days of diagnosis compared to TM patients (OR: 0.58; 95% CI: 0.34-0.97; P: 0.04). There were no statistical differences in starting treatment at 30 (OR: 0.79; 95% CI: 0.39-1.56; P: 0.5) or 45 days (OR: 0.87; 95% CI: 0.48-1.57; P: 0.7). Subgroup analysis indicated substantial delays in pembrolizumab initiation within 90 days among lowest SES (OR: 0.31; 95% CI: 0.13-0.71; P: 0.007; P-interaction: 0.087) and those treated in community hospitals (OR: 0.49; 95% CI: 0.27-0.87; P: 0.016; P-interaction: 0.095). Conclusions: Patients with TM were more likely to initiate pembrolizumab earlier than those with MA. This underscores necessity to scrutinize the influence of MA on cancer delivery, especially for patients with lower SES.

Baseline characteristics.

Variable TM MA P

Age

0.4

>=65 66 (50%) 40 (47%)

>=75 67 (50%) 45 (52%)

>=85 0 (0%) 1 (1%)

Race

1

NHW 97 (73%) 62 (72%)

Non-NHW 36 (27%) 24 (28%)

ECOG

0.7

0/1 95 (71%) 68 (79%)

2 18 (14%) 8 (9%)

3/4 8 (6%) 4 (5%)

Unknown 12 (9%) 6 (7%)

SES

0.04

0/1 38 (29%) 36 (42%)

3/4/5 88 (66%) 42 (49%)

Unknown 7 (5%) 8 (9%)

Diagnosis Year

0.5

2020 36 (27%) 22 (26%)

2021 41 (31%) 34 (40%)

2022 39 (29%) 18 (21%)

2023 17 (13%) 12 (14%)

DeNovo

0.7

Yes 80 (60%) 49 (57%)

No 53 (40%) 37 (43%)

Practice

0.4

Academic 16 (12%) 15 (17%)

Community 117 (88%) 71 (83%)

Time to Pembrolizumab Initiation (days)

OR (95CI)

P

30 1 0.79 (0.39-1.56) 0.50

45 1 0.87 (0.48-1.57) 0.66

90 1 0.58 (0.34-0.97) 0.04