# U-shaped prognostic signature: baseline platelet count and morphological parameters predict ovarian cancer outcomes in a 265-patient cohort

## TL;DR

This study finds that baseline platelet count and morphology can predict ovarian cancer recurrence in a U-shaped pattern, offering new insights into prognosis.

## Contribution

The novel U-shaped relationship between platelet count and ovarian cancer recurrence risk is identified, along with the modulating role of platelet morphology.

## Key findings

- Baseline platelet count shows a U-shaped association with recurrence risk, with lowest risk at 236×10⁹/L.
- Platelet morphological indicators have a protective effect in certain patient subgroups.
- The U-shaped relationship is confirmed using multinomial logistic regression and threshold analysis.

## Abstract

This study aims to explore the predictive value of baseline platelet count and its morphological indicators for the prognosis of ovarian cancer patients.

A retrospective cohort study was conducted at a gynecological oncology center in Beijing, involving ovarian cancer patients between 2011 and 2022 and followed up until December 2024. Data were extracted from clinical information system. The primary endpoints were recurrence; the primary indicator was progression-free survival during the follow-up period.

A total of 265 patients was included in this study. During the follow-up period, 110 patients recurred, whereas 155 patients achieved remission. Univariate analysis revealed that baseline platelet count was associated with progression-free survival. The stratified analysis presented a U-shaped curve by smooth curve fitting. Threshold effect analysis indicated that the inflection points of the U-shaped curve occurred at platelet count of 236×109/L (95% CI 222-256×109/L), the lowest risk of recurrence. The U-shaped curve was confirmed by Multinomial logistic regression (P < 0.005). The relationship between platelet morphological indicators and recurrence risk is modulated by the level of baseline platelet count(P < 0.05). In the third tertile of platelet distribution, morphological indicators are associated with recurrence risk and exhibit a protective effect; however, in patients with high-grade serous carcinoma and at different clinical stages, recurrence risk is not significantly associated with morphological indicators; but morphological indicators still have a protective effect on other pathological types (PDW, OR 0.6, P = 0.017; MPV, OR 0.3, P = 0.025; PLCR, OR 0.9, P = 0.022,respectively). The non-parametric Mann-Whitney U test also showed that the predictive value of baseline platelet morphology indicators for recurrence risk was only demonstrated in patients with other pathological types.

This study reveals a significant nonlinear association between platelet count and morphology and the risk of recurrence of ovarian cancer. Exploring the complex mechanisms linking baseline platelet characteristics to the prognosis of ovarian cancer will help facilitate the application of platelets as meaningful prognostic indicators and therapeutic targets in clinical practice.