Patterns of chemotherapy delivery following neuropathy in patients with stage I-IIIA breast cancer.

Abstract

e12514 Background: Chemotherapy-induced neuropathy is a common toxicity during treatment for breast cancer and may prompt changes in chemotherapy delivery, including a reduction in cumulative dose or dose intensity, which are associated with worse breast cancer outcomes. Real-world patterns of chemotherapy delivery following neuropathy and associated patient- and provider/facility-level factors remain poorly characterized. Methods: We analyzed data from the Optimal Breast Cancer Chemotherapy Dosing (OBCD) Study of women with primary stage I–IIIA breast cancer diagnosed between 2005-2019 who received adjuvant chemotherapy at Kaiser Permanente Northern California (KPNC) or Washington (KPWA) and developed chemotherapy-induced neuropathy during treatment. Neuropathy was identified using diagnostic codes and clinical documentation. The primary outcome was defined as chemotherapy modification in the cycle following neuropathy identification, including dose reduction ≥10%, regimen change, or chemotherapy discontinuation. Multivariable generalized linear models of the Poisson family were used to estimate adjusted prevalence ratios (aPRs) and 95% confidence intervals (CIs) for patient- and provider/facility-level factors (e.g., provider gender, practice size, treatment facility urbanicity (defined as 100% urban vs < 100% urban)) associated with the composite outcome. Provider/facility analyses were limited to KPNC sub-cohort. Results: Among 3,801 patients on chemotherapy who developed neuropathy, 12.9% experienced a chemotherapy modification in the subsequent cycle. The most common modification was dose reduction ≥10% (11.3%), followed by regimen change (2.4%), and early discontinuation (0.8%). The likelihood of modification increased with age (p-trend < 0.001), particularly among patients aged ≥75 years versus < 40 years (aPR 3.01; 95% CI 1.76–5.35). Other patient factors associated with any chemotherapy modification included HER2-positive disease (aPR 1.69; 95% CI 1.24–2.30), pre-existing liver disease (aPR 1.73; 95% CI 1.09–2.72), and receipt of dose-dense chemotherapy (aPR 1.69; 95% CI 1.24–2.30). Chemotherapy agent type was also associated with the likelihood of subsequent modification. Among provider/facility factors, treatment at non-urban facilities was associated with higher likelihood of chemotherapy modification (aPR 1.86; 95% CI 1.25-2.76). No other provider or facility factors were associated with chemotherapy modification. Conclusions: Chemotherapy delivery after neuropathy varied by patient, treatment, and facility characteristics, underscoring the complexity of treatment decisions following toxicity and the need to better characterize factors influencing chemotherapy delivery in real-world practice.