Outcomes and prognostic factors in patients with locally advanced cervical cancer treated with concurrent chemoradiotherapy

Abstract

Background: To investigate the prognostic factors affecting long-term survival in locally advanced cervical cancer (LACC) patients treated with concurrent chemoradiotherapy (CCRT).

Methods: We retrospectively analyzed 192 naive LACC (stage IIB-IVA) patients who underwent intensity-modulated radiotherapy (IMRT) with concurrent platinum-based chemotherapy in Xiangya Hospital from January 2014 to June 2017. The clinicopathological factors of all patients were collected. To explore the relationship between factors and prognosis, survival rates were estimated by the Kaplan-Meier method. Univariate and multivariate Cox proportional hazards models were used to evaluate the effect of various factors on overall survival (OS) and progression-free survival (PFS). The nomogram and calibration curves were generated on the basis of survival analysis.

Results: The median follow-up time was 39.5 months. There-year rates of OS and PFS were 89.1% and 82.8%. LACC patients with non-squamous cell carcinoma [NSCC, including adenocarcinoma or adenosquamous carcinoma (AC/ASC)], advanced stage (IIIA-IVA), initially positive lymph node (pelvic or para-aortic lymph node, PLN/PALN), and a lower pretreatment hemoglobin (HGB) level (< 126 g/L) had lower survival rates. In univariate analysis, patients with NSCC, advanced stage, PLN or PALN metastasis had worse OS. Patients with NSCC, advanced stage, PLN or PALN metastasis, and a lower pretreatment HGB level had worse PFS. In multivariate analysis, NSCC and PALN metastasis were independent prognostic parameters of OS. NSCC, PALN metastasis and a lower pretreatment HGB level were independent prognostic parameters of PFS.

Conclusions: NSCC and PALN metastasis were poor prognostic factors of OS and PFS, a lower pretreatment HGB level was an independent prognostic factor of PFS in LACC patients treated with CCRT.

Keywords: Concurrent chemoradiotherapy; Locally advanced cervical cancer; Prognostic factors; Recurrence or metastasis.

Fig. 1

Flow chart outlining the patient-collection process

Fig. 2

The OS and PFS rates according to Histology (a), FIGO Stage (b), PLN (c), PALN (d), Pretreatment HGB level (e). The differences in OS and PFS were estimated using the Kaplan–Meier method. OS: Overall survival; PFS: progression-free survival; SCC: squamous cell carcinoma; FIGO: International Federation of Gynecology and Obstetrics; PLN: pelvic lymph node; PALN: para-aortic lymph node; HGB: hemoglobin

Fig. 3

Forest plots based on univariate and multivariable Cox analyses of OS (a, b) and PFS (c, d) in patients with LACC. OS: Overall survival; PFS: progression-free survival; PLN: pelvic lymph node; PALN: para-aortic lymph node; HPV: human papillomavirus; ACT: adjuvant chemotherapy; HGB: hemoglobin

Fig. 4

Nomogram model for LACC patients to predict the 2- and 3-year OS. PLN indicates positive (1) or negative (0) pelvic lymph nodes. PALN indicates positive (1) or negative (0) para-aortic lymph nodes. To use, find patient’s age on age axis, then draw straight line upward to points axis to determine how many points patient receives for age and do this again for other variable axes. The total points predicted on the bottom scale was calculated by summing all points on the scale for each variable to estimate the probabilities of 2-year and 3-year OS rates by plotting a vertical line. OS: Overall survival; SCC: squamous cell carcinoma; AC: adenocarcinoma; ASC: adenosquamous carcinoma; PLN: pelvic lymph node; PALN: para-aortic lymph node; HGB: hemoglobin

Fig. 5

The calibration curves for predicting OS for LACC patients at 2 years (a) and 3 years (b) in the verification. Calibration of the nomogram for OS by comparing the predicted survival (plotted on the X-axis) with the actual survival (plotted on the Y-axis). OS: Overall survival