Day 8 monocyte and lymphocyte counts as a novel and early predictor for discontinuation in gynecologic cancer patients receiving paclitaxel and carboplatin therapy: a retrospective observational study

TL;DR

This study shows that checking blood cell counts on Day 8 can help predict if gynecologic cancer patients will need to stop chemotherapy due to low blood cell levels.

Contribution

Day 8 monocyte and lymphocyte counts are identified as novel early predictors for treatment discontinuation in gynecologic cancer patients.

Findings

Day 8 monocyte and lymphocyte counts were strong predictors of chemotherapy interruption due to myelosuppression.

A model combining pre- and post-treatment values had better predictive performance than pre-treatment-only models.

A pre-treatment neutrophil count of 2,795/µL was identified as a significant risk factor for treatment interruption.

Abstract

Maintaining optimal treatment intensity is essential for therapeutic efficacy in gynecological malignancies undergoing paclitaxel plus carboplatin (TC) chemotherapy; however, myelosuppression frequently results in treatment delay or interruption, which can compromise long-term outcomes. While previous research has predominantly focused on pre-treatment laboratory values and late-onset adverse events, there is a crucial gap in identifying early, actionable biomarkers. This study sought to evaluate the utility of post-treatment values measured at an early time point (Day 8) to facilitate timely and proactive supportive interventions.

This retrospective analysis included 242 patients who underwent TC chemotherapy. Patients who discontinued therapy for reasons other than myelosuppression (e.g., allergic reactions or patient choice) were excluded from the final analyses. The primary outcome was treatment interruption (delay or cessation) due to persistence of myelosuppression during the first cycle. Predictive modeling assessed baseline, day 8, and combined hematological factors (including neutrophil, monocyte, and lymphocyte counts, and their derived ratios). Predictive performance was evaluated using multivariate logistic regression and receiver operating characteristic (ROC) curve analyses.

The model incorporating both pre- and post-treatment laboratory values demonstrated superior predictive performance, evidenced by a significantly higher area under the ROC curve than the model utilizing only pre-treatment factors (0.82 vs. 0.73, P = 0.011). Decision tree analysis identified critical cutoff points for post-treatment laboratory values: monocyte counts below 51/µL and lymphocyte counts below 994/µL on day 8 after chemotherapy. Additionally, a pre-treatment neutrophil count of 2,795/µL was identified as a significant risk factor.

Early post-treatment hematological indices, particularly the day 8 monocyte and lymphocyte counts, were identified as more robust predictors of chemotherapy interruption due to myelosuppression, in addition to pre-treatment indices. These findings suggest that routine monitoring and proactive supportive interventions guided by these specific early post-treatment markers should be implemented in clinical pharmacy practice to enhance treatment continuity and optimize patient outcomes.

Figures