Factors associated with decreased treatment intensity in patients with metastatic colon cancer: A real world analysis.

Abstract

11168 Background: The frontline treatments for metastatic colon cancer (mCC) include intensive therapies (e.g., doublet or triplet chemotherapy backbones with biologic agents) and non-intensive therapies (e.g., chemotherapy monotherapy, biologic agents alone, or a combination thereof). Randomized trials have shown that non-intensive therapies offer shorter overall survival compared to intensive therapies; thus, they are recommended only for those unable to tolerate intensive therapy, such as those with poor performance status (PS). There is a need to identify which patients are receiving non-intensive therapies in the real world to ameliorate disparities in undertreatment. Methods: We conducted a retrospective cohort study by leveraging the nationwide Flatiron Health electronic health record-derived deidentified database. This database is longitudinal, comprising deidentified patient-level structured and unstructured data, curated via technology-enabled abstraction. Patients included presented with mCC during 2013-2024 and received one of the first line treatments listed in the National Comprehensive Cancer Network (NCCN) guidelines. Therapies were categorized as intensive or non-intensive based on NCCN guidelines. Frailty was defined as baseline ECOG PS≥2, and older age was defined as ≥65 years (yrs) at diagnosis. Multivariable logistic regression was performed between treatment intensity and patient characteristics (age, frailty, sex at birth, year of diagnosis, race, socioeconomic status (SES)) to estimate odds ratios and 95% confidence intervals. Age and frailty were combined into multiple variables given significant interaction term. Results: Among 21,588 patients included in this study, 83.3% received intensive first-line therapy. Receipt of non-intensive therapy was associated with female sex at birth and earlier year of diagnosis but not race or SES. Frailty and age were associated with treatment choice. Older non-frail patients had a stronger association with receipt of non-intensive therapy compared to younger frail patients. Conclusions: In addition to frailty, older age was a strong predictor of the receipt of non-intensive therapy among patients with mCC. More research and education around identifying frailty is needed to avoid undertreatment of older patients with mCC. Patient characteristics Number of patients Proportion receiving non-intensive therapies Odds ratio (95% confidence interval) P value < 65 yrs and non-frail 7432 7.50% Reference ≥ 65 yrs and frail 1547 37.0% 7.10 (6.20, 8.12) <0.01 ≥ 65 yrs and missing frailty 2596 28.0% 4.22 (3.72, 4.77) <0.01 ≥ 65 yrs and non-frail 6600 18.8% 2.82 (2.53, 3.13) <0.01 < 65 yrs and missing frailty 2544 16.0% 2.13 (1.86, 2.45) <0.01 < 65 yrs and frail 869 11.3% 1.50 (1.19, 1.88) <0.01 Male 11630 14.7% Reference Female 9958 19.1% 1.36 (1.26, 1.46) <0.01 Year of diagnosis 0.91 (0.90, 0.92) <0.01