Pace of Cervical Cancer Stage Progression Prior to Radiation Therapy Initiation, an Instrumental Variable Analysis.

Abstract

Delays in initiating radiation therapy for cervical cancer are common and may contribute to poorer outcomes. However, the pace of stage progression remains uncertain, limiting the development of quality targets and mitigation strategies. We conducted a prospective cohort study of women with cervical cancer treated with radiation in Botswana between 2012 and 2019. We used an instrumental variable (IV) approach based on seasonal variation in access to treatment to estimate the causal effect of treatment delay on cancer stage progression. The month of cancer diagnosis served as a natural source of variation, with women diagnosed during end-of-year holiday disruptions (November-February) experiencing longer delays on average than women at other times of the year. Using two-stage weighted models of mean delay and mean cancer stage at time of treatment, we estimated the average time required for cervical cancer stage to progress (eg, IIIA to IIIB). To contextualize findings we examined the association between cancer stage at treatment initiation and overall survival. Among 763 eligible participants, 494 were diagnosed during routine access periods and 269 during delayed periods. The mean time to treatment initiation was longer (IV partial F statistic 13.1) in the delayed access group (12.4 weeks) compared to the routine access group (11.2 weeks). The groups were otherwise well-balanced. In the primary analysis, cervical cancer stage progression was estimated to occur over a mean of 8.6 weeks (95% CI: 7.1-10.9). Modeled time to stage progression differed by HIV status (p <0.001) with estimated 7.6 weeks (95% CI: 6.4-9.2) for women with HIV and 8.7 weeks (95% CI: 6.1-15.3) for women without HIV. Stage increase was associated with 38% (95% CI: 33 to 44%) reduction in survival time. Modest treatment delays commonly encountered in routine care were associated with measurable cervical cancer progression likely decreasing survival. Initiating treatment within four weeks from diagnosis minimizes the risk of stage progression.