Radiotherapy can improve overall survival in patients with lymph-node positive, high-grade neuroendocrine cervical cancer: construction of two prognostic nomograms to predict treatment outcome

Abstract

Background: To explore the beneficial subgroups after radiotherapy in high-grade neuroendocrine cervical cancer (HGNECC) and construct two survival prognosis models to quantify the efficacy of radiotherapy assessment.

Methods: In this retrospective study, we included 592 eligible samples from the Surveillance, Epidemiology, and End Results (SEER) database and 56 patients with lymph-node positive HGNECC from Chongqing Medical University. Cox regression analysis was used to identify independent survival prognosis risk factors for HGNECC patients. Propensity score matching (PSM) was employed as it balances the baseline differences among grouping methods. Kaplan-Meier (K-M) curves were used to analyze survival differences among different groups. Two survival prediction nomograms were constructed separately (using the "rms" package in R software) based on whether radiotherapy was administered. The stability and accuracy of these models were assessed using receiver operating characteristic (ROC) curves and calibration curves in both the training and validation datasets. P<0.05 was considered to indicate statistically significant differences.

Results: Age, Federation of Gynecology and Obstetrics (FIGO)-stage, and treatment methods (surgery vs. chemotherapy) were independent risk factors that affected survival prognosis (P<0.05). Radiotherapy showed adverse effects on survival in patients with early tumor staging, lymph-node negative status, and absence of distant metastasis (all P<0.05). The lymph-node positive group had a beneficial response to radiotherapy (P<0.05), and patients with metastasis in the radiotherapy group showed a survival protection trend (P=0.069).

Conclusion: In HGNECC, patients with lymph-node positive status can benefit from radiotherapy in terms of survival outcomes. We constructed two survival prediction models based on whether radiotherapy was administered, thereby offering a more scientifically guided approach to clinical treatment planning by quantifying the radiotherapy efficacy.

Keywords: SEER database; high-grade neuroendocrine cervical cancer; nomogram; radiotherapy; survival prognosis.

Figure 1

The flowchart of enrolled patients screening. ^aCQMU, Chongqing Medical University. Here refers to the First and Second Affiliated Hospitals specifically.

Figure 2

Compares the survival differences between the R and NR groups before and after PSM in all patients in the SEER database. (A) Survival probability before PSM. (B) Survival probability after PSM. R, Radiotherapy; NR, Non radiation therapy.

Figure 3

Compare the survival difference before and after PSM between the R- and NR-group in different indicators: (A–D), grouping according to FIGO staging; (E–H), grouping according to lymph nodes status; (I–L), grouping according to whether distant metastasis; (M–P), grouping according to histological difference.

Figure 4

Constructing and validating the two nomograms based on therapeutic differences: (A) nomogram of patients without radiotherapy; (B) nomogram of patients with radiotherapy; (C–H) Receiver operating characteristic curve (ROC) of predictive models for 1- and 3- years in NR-group and R-group individually.