Developing a Pretreatment Risk Stratification Model: A Clinical Prediction Tool for Guiding Prognostic Grouping in Cervical Cancer Patients Undergoing Chemoradiotherapy

Abstract

Background:

Despite many reported pretreatment prognostic factors in cervical cancer patients, no integrated model has been established. This study aimed to evaluate the prognostic significance of established pretreatment factors in cervical cancer patients undergoing concurrent chemoradiotherapy (CCRT) and to develop a simple and practical model for pretreatment risk stratification.

Methods:

Fifty-one patients with cervical cancer treated with CCRT between September 2009 and July 2022 were retrospectively analyzed. The median follow-up period was 74.6 months, and the median patient age was 58 years. Clinicopathological and hematological factors, including age, body mass index (BMI), pathology, hemoglobin (Hgb), neutrophil-to-lymphocyte ratio (NLR), platelet-to-lymphocyte ratio (PLR), squamous cell carcinoma antigen (SCC-Ag), and International Federation of Gynecology and Obstetrics (FIGO) stage, were collected. Radiological and metabolic factors were assessed using magnetic resonance imaging (MRI), computed tomography (CT), and fluorine-18-fluorodeoxyglucose positron emission tomography/computed tomography (F-18 FDG PET/CT), from which the primary tumor volume (pTV) and the maximum standardized uptake value (SUV_max) were measured. The prognostic significance of factors for progression-free survival (PFS) and overall survival (OS) was evaluated using the Cox proportional hazards models.

Results:

The 5-year OS and PFS rates were 75.5% and 72.6%, respectively. In univariate analyses, BMI (p = 0.020) and FIGO stage (p < 0.001) were associated with PFS, while age, pathology, Hgb, NLR, PLR, SCC-Ag, pTV, and SUV_max were not. Multivariate analysis identified FIGO stage as the only independent prognostic factor for PFS (hazard ratio [HR]: 4.385; 95% CI: 1.865–10.310; p < 0.001). For OS, Hgb (p = 0.044), BMI (p = 0.024, and FIGO stage (p < 0.001) were significant in univariate analyses, whereas BMI (HR: 3.207; 95% CI: 1.157–8.893; p = 0.025), and FIGO stage (HR: 3.604; 95% CI: 1.559–8.334; p = 0.003) remained significant in the multivariate analysis. The optimal BMI cut-off, determined by the receiver operating characteristic (ROC) analysis, was 21.2 kg/m².

Conclusions:

FIGO stage and BMI were the most influential pretreatment factors associated with survival in cervical cancer patients undergoing CCRT. These findings support a straightforward pretreatment risk-stratification approach based on readily obtainable information, intended to assist baseline risk communication rather than guide treatment decisions. External validation in larger, multicenter cohorts is needed to confirm its clinical reliability, but these findings suggest a practical approach to identifying patients with distinct prognostic risks before treatment.

Graphical abstract

Keywords

cervical cancer / chemoradiotherapy / prognostic factor

1. Introduction

Cervical cancer remains one of the most common malignancies in the world and is still associated with a high risk of death. According to data from the International Cancer Research Institute and the World Health Organization, 604,000 new cervical cancer cases were diagnosed in 2020, resulting in 342,000 deaths [1, 2]. The cause of cervical cancer is strongly related to persistent human papillomavirus (HPV) infection, particularly HPV-16 and HPV-18, which together account for the most cases of invasive disease [3]. Early detection of cervical cancer commonly starts with Pap smear screening, followed by biopsy for confirmation and imaging studies for staging [4]. However, despite improvements in screening and HPV vaccination, many patients are still diagnosed at locally advanced stages, indicating that reliable pretreatment markers that can guide clinical decisions are still needed [5] and that cervical cancer remains a major cause of death. When surgery is not an option—often because the disease is too advanced or the patient is not a suitable surgical candidate—the main treatment strategy shifts to concurrent chemoradiotherapy (CCRT). The definitive radiotherapy (RT) approach typically starts with external beam radiation therapy (EBRT) delivered to the pelvis and is then followed by intracavitary brachytherapy (ICR) to boost the dose directly to the tumor while minimizing exposure to the surrounding healthy tissues [6, 7]. When assessing the prognosis of patients with cervical cancer treated with RT, several pretreatment factors are already known to be important. These can be grouped into a few broad categories that clinicians commonly evaluate before starting treatment. The clinicopathological factors included age, body mass index (BMI), clinical stage, and pathology. The hematological and radiological factors, including hemoglobin (Hgb), hematocrit (Hct), neutrophil-to-lymphocyte ratio (NLR), platelet-to-lymphocyte ratio (PLR), squamous cell carcinoma antigen (SCC-Ag), primary tumor volume (pTV) or size, are also routinely evaluated. In addition, metabolic factors, including the maximum standardized uptake value (SUV_max) in fluorine-18-fluorodeoxyglucose positron emission tomography/computed tomography (F-18 FDG PET/CT) and, when available, metabolic tumor volume (MTV) and total lesion glycolysis (TLG), provide further insight into the overall disease burden [8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19]. Despite such well-known pretreatment clinicopathological, hematological, radiological, and metabolic factors, their relative prognostic value has not been systematically compared, and no simple model integrating these factors has been established. Therefore, this study aimed to evaluate the prognostic significance of major and readily obtainable pretreatment factors and to develop a practical, clinically applicable risk-prediction score for cervical cancer patients undergoing CCRT.

2. Materials and Methods

2.1 Patient Selection

A total of 51 patients with histologically verified uterine cervical cancer underwent RT at our institution between September 2009 and July 2022. The patients in this study were staged according to the 2018 International Federation of Gynecology and Obstetrics (FIGO) staging system [20]. The study inclusion criteria were: (a) age \ge18 years, (b) pathologically confirmed SCC or adenocarcinoma (ADC), (c) FIGO stage IB-IVA without evidence of distant metastasis, (d) Eastern Cooperative Oncology Group performance status of 0–2, (e) receipt of EBRT with concurrent chemotherapy, followed by ICR. The study exclusion criteria were: (a) incomplete clinical records, (b) insufficient or missing pretreatment laboratory or imaging data, (c) incomplete RT, (d) received RT as palliative treatment, (e) prior surgical or therapeutic intervention before CCRT, (f) a previous malignancy in another organ, and (g) evidence of active infection or inflammatory disease (e.g., pneumonia or inflammatory joint disease), or hematologic disorders prior to CCRT.

Clinical factors collected included BMI, pathology, Hgb, NLR, PLR, SCC-Ag, and FIGO stage. Radiological and metabolic factors were evaluated using magnetic resonance imaging (MRI), CT, and PET/CT. The pTV was recorded as the radiological factor, and the SUV_max of the primary lesion was used as the metabolic factor.

The BMI and pTV were calculated using the following formulas:

\text{BMI}=\text{weight}(\mathrm{kg})/{[\text{height}(\mathrm{m})]}^2

\mathrm{pTV}=\mathrm{L}\times \mathrm{W}\times \mathrm{H}\times (\pi /6)

where L, W, and H represent the maximal tumor length, width, and height, respectively.

2.2 Treatment

All patients first underwent pelvic EBRT delivered with 10- or 15-megavoltage (MV) photon beams using a four-field box technique. Total EBRT doses ranged from 50.4 to 84.9 Gy (median, 59.4 Gy), administered in daily fractions of 1.8–2 Gy, five days per week. Concurrent chemotherapy was given during EBRT for all patients; one patient received weekly carboplatin (290 mg/m²), whereas the remaining 50 patients received weekly cisplatin (40–60 mg/m²). In accordance with international guidelines [21, 22], cisplatin-based CCRT was used as the standard definitive treatment for all patients with FIGO stages IB2–IVA. Since no patient underwent primary surgery, stage-based variations in surgical management did not apply to this cohort.

After completing a median EBRT dose of 45 Gy, high-dose-rate (HDR) ICR was initiated using a Microselectron HDR system (Ir-192 source; Nucletron, Veenendaal, The Netherlands). Before each ICR session, an MRI or CT was performed to evaluate the most recent tumor response. If marked tumor reduction was observed or if the cumulative EBRT dose indicated that organ at risk (OAR) limits for the rectum, bladder, or bowel could be reached, the ICR fraction dose was adjusted accordingly to remain within safe tolerance levels. These dose modifications were made on an individual basis as part of routine clinical practice. A total HDR-ICR dose of 20–32 Gy (median, 24 Gy) was delivered twice weekly in fractions of 3–5 Gy. No patient received consolidation chemotherapy after completion of CCRT. The primary endpoint included both treatment response and treatment failure, which were assessed by gynecological examination, Pap smear test, cross-sectional imaging (CT or MRI from chest to pelvis), and PET/CT when available.

2.3 Statistical Analysis

All statistical analyses were conducted using R software (version 4.2.1; R Foundation for Statistical Computing, Vienna, Austria). Categorical variables were reported as counts and percentages. For continuous variables, we checked normality with the Shapiro-Wilk test and presented them as median and interquartile range. Progression-free survival (PFS) and overall survival (OS) were estimated using the Kaplan–Meier method, with group comparisons performed via the log-rank test. Continuous variables, including BMI, Hgb, NLR, PLR, SCC-Ag, pTV, and SUV_max, were dichotomized according to optimal cut-off values identified through receiver operating characteristic (ROC) curve analysis. The ROC procedure identified thresholds that maximized sensitivity and specificity, and these cut-off points were subsequently applied for classification in the survival analyses. For the PFS analysis, disease recurrence—either locoregional or distant—was defined as the endpoint, whereas death was considered the endpoint for the OS analysis. In this study, univariate analyses were performed using the log-rank test. For the multivariate analysis, prognostic variables were evaluated using Cox proportional hazards regression. Variables with p 0.10 in the univariate analysis were initially considered, and the final multivariable model was selected using a stepwise procedure based on the Akaike Information Criterion (AIC), which identified the model with the best balance between goodness of fit and parsimony. Hazard ratios (HR) and 95% CI were reported, and statistical significance was defined as p 0.05. A risk-based classification model was generated based on \beta-coefficients from the multivariate analysis. \beta-coefficients were obtained directly from the multivariable Cox proportional hazards model, which estimates the log-hazard for each prognostic factor. HRs were derived as the exponential transformation of these coefficients [HR = exp(\beta)]. The \beta-coefficients from the final multivariable model were used as the basis for assigning points in the risk-score calculation. Model calibration and discriminatory performance were internally validated through bootstrap resampling with 400 replicates (B = 400), which generated bias-corrected estimates for assessing model stability. This approach allowed evaluation of potential overfitting by comparing apparent and bias-adjusted calibration curves. The bootstrap procedure further ensured robustness by repeatedly re-estimating the prognostic model across resampled datasets.

3. Results

The baseline characteristics of the 51 patients are presented in Table 1. The Shapiro–Wilk test results for all continuous variables are provided in Supplementary Table 1. The median age was 58 years (range, 37–81 years), and 84.3% of patients were younger than 75 years. Most patients (72.5%) had FIGO stage III or IV disease, and all underwent CCRT.

The median follow-up duration was 74.6 months (range, 9.9–156.4 months). The median age of the cohort was 58 years (range: 37–81 years). The 5-year OS and PFS outcomes were 75.5% and 72.6%, respectively (Fig. 1).

3.1 PFS Analysis

In the univariate analysis, age (p = 0.202), pathology (p = 0.843), Hgb (p = 0.166), NLR (p = 0.210), PLR (p = 0.552), SCC-Ag (p = 0.444), pTV (p = 0.435), and SUV_max (p = 0.342) were not associated with PFS. In contrast, BMI (p = 0.020) and FIGO stage (p 0.001) were associated with PFS (Table 2). According to the multivariate analysis, FIGO stage (HR: 4.385; 95% CI: 1.865–10.310; p 0.001) was a significant prognostic factor for PFS (Fig. 2) (Table 3).

3.2 OS Analysis

Age (p = 0.284), pathology (p = 0.689), NLR (p = 0.399), PLR (p = 0.470), SCC-Ag (p = 0.231), pTV (p = 0.188), and SUV_max (p = 0.269) were not correlated, while Hgb (p = 0.044), BMI (p = 0.024), and FIGO stage (p 0.001) were correlated (Table 2). In the multivariate analysis, BMI (HR: 3.207; 95% CI: 1.157–8.893 p = 0.025), and FIGO stage (HR: 3.604; 95% CI: 1.559–8.334; p = 0.003) showed prognostic significance for OS (Fig. 3) (Table 3). As BMI was the only continuous variable significantly associated with OS, we performed an ROC analysis to determine its optimal cut-off value. The ROC curve for BMI is presented in Supplementary Fig. 1.

The results of the univariate analyses for PFS and OS of all factors are summarized in Table 2. A summary of significant and non-significant prognostic factors for both PFS and OS is provided in Supplementary Table 2.

3.3 Risk Stratification Model Based on Prognostic Scoring

A prognostic risk score for PFS was developed based on the \beta-coefficients of the risk factors. In this scoring model applied for PFS, FIGO stage (I, II, III, or IV) was incorporated as a prognostic variable, with points assigned based on stage. FIGO stages I–III were allocated a point of 0, whereas FIGO stage IV was allocated a point of 14. Risk stratification for PFS was categorized as follows: low-risk group (point = 0; n = 45, 88.2%), high-risk group (point = 14; n = 6, 11.8%). The two groups showed a significant difference in 5-year PFS (79.5% [95% CI: 64.20%–88.80%] vs. 16.7% [95% CI: 0.77%–51.70%], p 0.001) (Fig. 4). The prognostic risk score for OS was constructed using the following prognostic factors, with weights assigned accordingly: BMI 21.2 kg/m² (assigned a point of 11), and FIGO stage I, II, III, or IV (assigned a point of 0 for stage I–III and 12 for stage IV).

OS risk stratification resulted in three groups: low-risk group (point = 0, n = 31, 60.8%), intermediate-risk group (point = 11, n = 17, 33.3%), and high-risk group (point = 23, n = 3, 5.9%). A statistically significant differences in the 5-year OS rates was found across the three groups: 93.2% [95% CI: 75.7% to 98.3%] for low-risk group, 57.8% [95% CI: 31.1% to 77.3%] for intermediate-risk group, and 0% [95% CI: NA] for high-risk group p 0.001) (Fig. 4).

Table 3 summarizes the multivariate analysis results for all factors related to PFS and OS, along with the resulting risk stratification scoring model.

4. Discussion

In our study, BMI and FIGO stage were significant factors for survival outcomes among patients receiving CCRT for cervical cancer. In our study, the prognostic impact of FIGO stage and BMI was clearly demonstrated by the observed survival differences. Patients with stage I–III had a 5-year PFS of 79.5% and a 5-year OS of 93.2%, whereas those with stage IV showed markedly poorer outcomes, with both 5-year PFS and OS at 16.7%. BMI also showed a significant association with OS, with patients in the higher BMI group achieving a 5-year OS of 87.7% compared with 51.3% in the lower BMI group. Regarding radiotherapy, there was a wide EBRT dose range; however, the higher doses of 84.9 Gy and 70.4 Gy were given only to two patients with bulky pelvic nodal disease, for whom escalation was clinically appropriate while respecting OAR constraints. All other patients received EBRT within the standard range of 50.4–66.6 Gy, and no toxicities were observed in those treated with higher doses. Because all patients subsequently underwent HDR brachytherapy, the overall biological dose to the cervix remained relatively consistent, making it unlikely that these dose variations influenced the results of the study.

Previous study has investigated the relationship between BMI and survival outcomes in gynecological cancers. While obesity has traditionally been considered a negative prognostic factor, recent systematic reviews and meta-analyses have shown that a high BMI does not necessarily confer a survival benefit, and in some cases may even be associated with worse overall survival [14]. Additionally, a recent study have also shown favorable outcomes in higher BMI groups; for example, overweight patients demonstrated significantly better survival than both underweight and obese patients [23]. Furthermore, a qualitative study found that some overweight patients reported better tolerance to therapy, especially RT [24]. In another study, cachexia risk stratification using BMI showed shorter survival (median 4.83 months) in the low BMI groups [25]. Consistent with the studies mentioned above, in our study, BMI was significant for PFS and also for OS. The reasons why BMI might be linked to survival prognosis could be explained as follows: nutritional reserves could buffer against treatment-related cachexia [26], altered pharmacokinetics can lead to improved drug tolerance [27, 28], earlier detection due to more frequent screening could be performed in some regions [29], hormonal milieu possibly affects tumor growth dynamics differently [30], and sarcopenia, rather than BMI alone, serves as a more decisive prognostic factor for OS. Patients with higher BMI and preserved muscle mass tend to exhibit more favorable clinical outcomes [28, 31].

In this study, prognostic factor analysis was performed on patients classified according to the 2018 FIGO staging system. Due to the limited number of patients, the FIGO stages were divided into four groups (I, II, III, and IV) for analytical purposes. When FIGO stage was dichotomized (I–III vs. IV), the prognostic significance remained unchanged, further supporting the robustness of our findings. The results confirmed that the FIGO stage was a statistically significant prognostic factor for both OS and PFS. The 2018 revision of the FIGO staging system, which incorporates imaging and nodal status, is widely recognized as a major advance in predicting cancer outcomes [20, 32]. Studies have demonstrated that these changes allow for earlier intervention, more accurate risk stratification, and tailored treatment, all of which improve outcomes [33, 34, 35]. Overall, these results highlight the complementary value of BMI and FIGO stage as pretreatment prognostic markers. Because our study period included years before the introduction of the 2018 FIGO system, earlier cases were reclassified according to the updated criteria to maintain consistent staging. Since no patient underwent surgery, and the change in staging did not influence the CCRT approach, this reclassification is unlikely to have affected clinical management or the validity of the prognostic analyses. FIGO stage remained the most influential factor, reflecting the overall extent of disease. BMI provided additional insight into treatment tolerance and general health, with lower BMI linked to poorer survival. Because both measures are simple to obtain and represent different aspects of the patient’s status, they offer practical utility in estimating prognosis and can support a straightforward, clinically useful risk-stratification model.

In the present study, hematological factors were not identified as a significant prognostic factor for survival. The prognosis of cervical cancer is significantly affected by anemia, which is commonly indicated by low Hgb and Hct levels. Numerous studies have confirmed that pretreatment anemia correlates with poor tumor oxygenation, leading to reduced radiotherapy efficacy, increased tumor hypoxia, and shorter overall and disease-free survival [11, 12, 36]. These results underscore the clinical importance of monitoring and managing anemia both prior to and during treatment to optimize patient outcomes. In the present study, Hgb demonstrated a significant association with OS in the univariate analysis. In the multivariate analysis, it showed a marginally significant association (HR: 2.366; 95% CI: 0.8578–6.5260; p = 0.096). The multivariate Cox analysis yielded a hazard ratio of 2.366, indicating more than a twofold increased risk of mortality associated with this factor. With a larger cohort, the confidence interval would be expected to narrow, and the observed trend might reach conventional levels of statistical significance. Furthermore, although NLR, PLR, and SCC-Ag have been previously reported as significant prognostic indicators for survival, they were not identified as independent prognostic factors in the multivariate analysis of the present study. Consistent with the findings of the present study, in some studies, NLR, PLR, and SCC-Ag have shown statistical significance in the univariate analyses; however, they were not identified as independent prognostic factors for survival in multivariate analyses [37, 38]. These findings were consistent with the results of the present study and suggested that the interactions among multiple variables should be considered in prognostic evaluations.

While the literature on cervical cancer and age as a prognostic factor is mixed, a study has suggested that advanced age may be an independent negative prognostic factor for mortality [15], whereas others have indicated no significant age-related differences in survival outcomes [8, 39]. Some studies have suggested that younger women may have a higher risk of local recurrence; however, unlike older women, they do not experience worsening outcomes due to recurrence [40, 41]. In our study, age did not have a significant effect on the survival outcomes.

In the present study, prognostic analyses were conducted using pTV as the radiological factor and SUV_max as the metabolic factor. Both factors lacked statistical significance in predicting survival outcomes. Consistent with our findings, other studies consistently reported that pTV was not a statistically significant predictor of survival [13, 16]. While previous studies have demonstrated that the tumor volume reduction rate (TVRR) during treatment is a significant prognostic factor for survival, calculating TVRR requires imaging obtained during radiotherapy, such as CT or MRI [13, 16]. Because the present study aimed to evaluate only pretreatment prognostic factors, TVRR was not included in our analysis. For the same reason, MTV, which also requires post-baseline imaging data, was not incorporated into this study. Yoo et al. [42] reported that SUV_max and average SUV were not independent prognostic indicators for survival. Although a correlation was observed between tumor size and SUV_max, multivariate analysis revealed no direct association with survival. Similarly, several other studies have demonstrated that SUV_max is not significantly associated with survival in multivariate systems and shows only a borderline relationship with OS, indicating that it is not an independent prognostic factor [10, 43]. Although metabolic prognostic indicators, such as MTV and TLG, which reflect the total metabolic burden, have been reported to demonstrate greater prognostic value than SUV_max [44, 45], they are more complex and difficult to measure in routine clinical practice. Therefore, this study included only the SUV_max, which is widely measured and used in most hospitals. In our analysis, radiological and metabolic factors such as pTV and SUV_max were evaluated, but did not show independent prognostic value. This suggests that patient-related systemic factors may exert a greater influence on outcomes than tumor-based measurements. The limited independent significance of pTV and SUV_max appears to reflect their strong correlation with FIGO stage, which already incorporates major elements of tumor burden and nodal involvement. When analyzed with the FIGO stage, much of their prognostic contribution was absorbed by the stage variable. Although SUV_max is less informative than volumetric PET factors such as MTV or TLG, these measures could not be assessed because the required imaging data were unavailable. Nevertheless, SUV_max is widely reported and routinely used in clinical practice, making its inclusion as a pretreatment metabolic factor reasonable in settings where volumetric PET metrics are not accessible. The results of our multivariate analysis using all these factors confirmed that BMI and the FIGO stage were significantly associated with survival. Based on these results, we developed a risk-score stratification model. Although FIGO staging requires assessment through imaging studies, other factors, such as BMI, can be readily obtained in clinical settings, suggesting that this classification model is easily feasible and can be applied to guide treatment decisions. One of the strengths of this study is that it identifies pretreatment prognostic factors that are clinically relevant in routine practice. Building on these findings, we developed a risk-score classification model intended to aid clinicians in planning treatment. To the best of our knowledge, no previous study has integrated multiple established pretreatment prognostic factors into a single predictive model for cervical cancer. By consolidating these factors into a unified framework, this study provides a structured approach that may support more consistent and informed patient assessment before therapy.

Limitations

This study has several limitations. The number of patients who met the eligibility criteria was modest, which limits the strength of the conclusions. Even so, the cohort was sufficient for the scope of the analysis because only a small number of variables were included in the multivariable Cox proportional hazards model, the events-per-variable ratio remained acceptable, and all patients were treated with a consistent CCRT protocol. However, the single-institution nature of the study suggests that the results should be interpreted with some caution. Accordingly, the prognostic tool derived from this study is intended solely to facilitate baseline risk communication and should not be used to inform or direct treatment decisions. Overall, the findings are best viewed as exploratory and will need to be confirmed in larger studies involving multiple centers.

5. Conclusions

In this study, FIGO stage and BMI were the most influential pretreatment factors for survival in patients receiving CCRT for cervical cancer. Their independent prognostic value offers a practical basis for a straightforward risk-stratification approach that may help clinicians recognize higher-risk patients and make more informed treatment decisions. Because both factors are routinely assessed before therapy, they provide a useful means of estimating prognosis at the time of diagnosis. This tool is intended solely to assist with baseline risk communication, not to guide treatment decisions. Although the model shows promise, its clinical reliability will need to be confirmed through external validation in multicenter studies and in larger patient cohorts. These findings provide a sound basis for future work to develop dependable pretreatment prognostic tools.

Availability of Data and Materials

The datasets analyzed during the current study are available from the corresponding author on reasonable request.

References

[1] Bray F, Ferlay J, Soerjomataram I, Siegel RL, Torre LA, Jemal A. Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA: a Cancer Journal for Clinicians. 2018; 68: 394–424. https://doi.org/10.3322/caac.21492.

[2] Sung H, Ferlay J, Siegel RL, Laversanne M, Soerjomataram I, Jemal A, et al. Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries. CA: a Cancer Journal for Clinicians. 2021; 71: 209–249. https://doi.org/10.3322/caac.21660.

[3] Cai H, Feng Y, Fan P, Guo Y, Kuerban G, Chang C, et al. HPV16 E6-specific T cell response and HLA-A alleles are related to the prognosis of patients with cervical cancer. Infectious Agents and Cancer. 2021; 16: 61. https://doi.org/10.1186/s13027-021-00395-y.

[4] Sachan PL, Singh M, Patel ML, Sachan R. A Study on Cervical Cancer Screening Using Pap Smear Test and Clinical Correlation. Asia-Pacific Journal of Oncology Nursing. 2018; 5: 337–341. https://doi.org/10.4103/apjon.apjon_15_18.

[5] Choi S, Ismail A, Pappas-Gogos G, Boussios S. HPV and Cervical Cancer: A Review of Epidemiology and Screening Uptake in the UK. Pathogens (Basel, Switzerland). 2023; 12: 298. https://doi.org/10.3390/pathogens12020298.

[6] McCormack M, Eminowicz G, Gallardo D, Diez P, Farrelly L, Kent C, et al. Induction chemotherapy followed by standard chemoradiotherapy versus standard chemoradiotherapy alone in patients with locally advanced cervical cancer (GCIG INTERLACE): an international, multicentre, randomised phase 3 trial. Lancet (London, England). 2024; 404: 1525–1535. https://doi.org/10.1016/S0140-6736(24)01438-7.

[7] Rose PG, Bundy BN, Watkins EB, Thigpen JT, Deppe G, Maiman MA, et al. Concurrent cisplatin-based radiotherapy and chemotherapy for locally advanced cervical cancer. The New England Journal of Medicine. 1999; 340: 1144–1153. https://doi.org/10.1056/NEJM199904153401502.

[8] Accorsi GS, Zanon JR, Santos MHD, Ubinha ACF, Schmidt R, Moretti-Marques R, et al. Cervical cancer in young women: Does age impact survival in cervical cancer? European Journal of Obstetrics, Gynecology, and Reproductive Biology. 2025; 305: 67–74. https://doi.org/10.1016/j.ejogrb.2024.12.003.

[9] Chen W, Wang R, Wu J, Wu Y, Xiao L. Comparison of surgical and oncological outcomes between different surgical approaches for overweight or obese cervical cancer patients. Journal of Robotic Surgery. 2024; 18: 107. https://doi.org/10.1007/s11701-024-01863-4.

[10] Chung HH, Kim JW, Han KH, Eo JS, Kang KW, Park NH, et al. Prognostic value of metabolic tumor volume measured by FDG-PET/CT in patients with cervical cancer. Gynecologic Oncology. 2011; 120: 270–274. https://doi.org/10.1016/j.ygyno.2010.11.002.

[11] Fyles AW, Milosevic M, Pintilie M, Syed A, Hill RP. Anemia, hypoxia and transfusion in patients with cervix cancer: a review. Radiotherapy and Oncology: Journal of the European Society for Therapeutic Radiology and Oncology. 2000; 57: 13–19. https://doi.org/10.1016/s0167-8140(00)00245-0.

[12] Grogan M, Thomas GM, Melamed I, Wong FL, Pearcey RG, Joseph PK, et al. The importance of hemoglobin levels during radiotherapy for carcinoma of the cervix. Cancer. 1999; 86: 1528–1536. https://doi.org/10.1002/(sici)1097-0142(19991015)86:8<1528::aid-cncr20>3.0.co;2-e.

[13] Lee JH, Lee SW, Kim JR, Kim YS, Yoon MS, Jeong S, et al. Tumour size, volume, and marker expression during radiation therapy can predict survival of cervical cancer patients: a multi-institutional retrospective analysis of KROG 16-01. Gynecologic Oncology. 2017; 147: 577–584. https://doi.org/10.1016/j.ygyno.2017.09.036.

[14] Pavone M, Goglia M, Taliento C, Lecointre L, Bizzarri N, Fanfani F, et al. Obesity paradox: is a high body mass index positively influencing survival outcomes in gynecological cancers? A systematic review and meta-analysis. International Journal of Gynecological Cancer: Official Journal of the International Gynecological Cancer Society. 2024; 34: 1253–1262. https://doi.org/10.1136/ijgc-2023-005252.

[15] Quinn BA, Deng X, Colton A, Bandyopadhyay D, Carter JS, Fields EC. Increasing age predicts poor cervical cancer prognosis with subsequent effect on treatment and overall survival. Brachytherapy. 2019; 18: 29–37. https://doi.org/10.1016/j.brachy.2018.08.016.

[16] Sun C, Wang S, Ye W, Wang R, Tan M, Zhang H, et al. The Prognostic Value of Tumor Size, Volume and Tumor Volume Reduction Rate During Concurrent Chemoradiotherapy in Patients With Cervical Cancer. Frontiers in Oncology. 2022; 12: 934110. https://doi.org/10.3389/fonc.2022.934110.

[17] Yu J, Huang L, Dong T, Cao L. Prediction of outcomes after chemoradiotherapy for cervical cancer by neutrophil-to-lymphocyte ratio and platelet-to-lymphocyte ratio. Journal of Obstetrics and Gynaecology: the Journal of the Institute of Obstetrics and Gynaecology. 2024; 44: 2361858. https://doi.org/10.1080/01443615.2024.2361858.

[18] Jin L, Cao F, Zhang Y, Dang Y, Wang F. Nomograms for predicting prognostic value of combined neutrophil-to-lymphocyte ratio and SCC-Ag in locally advanced cervical cancer. Translational Cancer Research. 2024; 13: 1323–1335. https://doi.org/10.21037/tcr-23-1501.

[19] Zhu M, Feng M, He F, Han B, Ma K, Zeng X, et al. Pretreatment neutrophil-lymphocyte and platelet-lymphocyte ratio predict clinical outcome and prognosis for cervical Cancer. Clinica Chimica Acta; International Journal of Clinical Chemistry. 2018; 483: 296–302. https://doi.org/10.1016/j.cca.2018.05.025.

[20] Wright JD, Matsuo K, Huang Y, Tergas AI, Hou JY, Khoury-Collado F, et al. Prognostic Performance of the 2018 International Federation of Gynecology and Obstetrics Cervical Cancer Staging Guidelines. Obstetrics and Gynecology. 2019; 134: 49–57. https://doi.org/10.1097/AOG.0000000000003311.

[21] Chino J, Annunziata CM, Beriwal S, Bradfield L, Erickson BA, Fields EC, et al. Radiation Therapy for Cervical Cancer: Executive Summary of an ASTRO Clinical Practice Guideline. Practical Radiation Oncology. 2020; 10: 220–234. https://doi.org/10.1016/j.prro.2020.04.002.

[22] Todo Y, Watari H. Concurrent chemoradiotherapy for cervical cancer: background including evidence-based data, pitfalls of the data, limitation of treatment in certain groups. Chinese Journal of Cancer Research = Chung-kuo Yen Cheng Yen Chiu. 2016; 28: 221–227. https://doi.org/10.21147/j.issn.1000-9604.2016.02.10.

[23] Tsai SY, Tsai MC, Hsu MS, Tsai LW, Hsu HC, Jhuang JR, et al. The association of different body weight classes and survival outcomes in patients with cervical cancer. Cancer Epidemiology. 2025; 96: 102801. https://doi.org/10.1016/j.canep.2025.102801.

[24] Sorensen JL, West MM, Robinson KM, Charlton ME, Lizarraga IM, Nash SH. Patient Perspectives on Impact of Weight and Weight Stigma on Breast and Cervical Cancer Treatment: A Qualitative Study. Cancer Medicine. 2025; 14: e70823. https://doi.org/10.1002/cam4.70823.

[25] Hu C, Cong M, Song C, Xu H, Guo Z, Zhou F, et al. A Novel Definition and Grading Diagnostic Criteria for Tumour-Type-Specific Comprehensive Cachexia Risk. Journal of Cachexia, Sarcopenia and Muscle. 2025; 16: e13744. https://doi.org/10.1002/jcsm.13744.

[26] Dunne RF, Loh KP, Williams GR, Jatoi A, Mustian KM, Mohile SG. Cachexia and Sarcopenia in Older Adults with Cancer: A Comprehensive Review. Cancers. 2019; 11: 1861. https://doi.org/10.3390/cancers11121861.

[27] da Fonseca GWP, Sato R, de Nazaré Nunes Alves MJ, von Haehling S. Current advancements in pharmacotherapy for cancer cachexia. Expert Opinion on Pharmacotherapy. 2023; 24: 629–639. https://doi.org/10.1080/14656566.2023.2194489.

[28] Prado CMM, Maia YLM, Ormsbee M, Sawyer MB, Baracos VE. Assessment of nutritional status in cancer–the relationship between body composition and pharmacokinetics. Anti-cancer Agents in Medicinal Chemistry. 2013; 13: 1197–1203. https://doi.org/10.2174/18715206113139990322.

[29] Tao W, Lagergren J. Clinical management of obese patients with cancer. Nature Reviews. Clinical Oncology. 2013; 10: 519–533. https://doi.org/10.1038/nrclinonc.2013.120.

[30] Tisdale MJ. Mechanisms of cancer cachexia. Physiological Reviews. 2009; 89: 381–410. https://doi.org/10.1152/physrev.00016.2008.

[31] Davis MP, Panikkar R. Sarcopenia associated with chemotherapy and targeted agents for cancer therapy. Annals of Palliative Medicine. 2019; 8: 86–101. https://doi.org/10.21037/apm.2018.08.02.

[32] Meng Q, Wang W, Liu X, Wang D, Zhang F. Nomograms Predicting Survival of Cervical Cancer Patients Treated With Concurrent Chemoradiotherapy Based on the 2018 FIGO Staging System. Frontiers in Oncology. 2022; 12: 870670. https://doi.org/10.3389/fonc.2022.870670.

[33] Brodeur MN, Dejean R, Beauchemin MC, Samouëlian V, Cormier B, Bacha OM, et al. Oncologic outcomes in the era of modern radiation therapy using FIGO 2018 staging system for cervical cancer. Gynecologic Oncology. 2021; 162: 277–283. https://doi.org/10.1016/j.ygyno.2021.05.023.

[34] Lee SI, Atri M. 2018 FIGO Staging System for Uterine Cervical Cancer: Enter Cross-sectional Imaging. Radiology. 2019; 292: 15–24. https://doi.org/10.1148/radiol.2019190088.

[35] Mohamud A, Høgdall C, Schnack T. Prognostic value of the 2018 FIGO staging system for cervical cancer. Gynecologic Oncology. 2022; 165: 506–513. https://doi.org/10.1016/j.ygyno.2022.02.017.

[36] Caro JJ, Salas M, Ward A, Goss G. Anemia as an independent prognostic factor for survival in patients with cancer: a systemic, quantitative review. Cancer. 2001; 91: 2214–2221.

[37] Jia SZ, Yang XJ, Yang D, Wang R, Yang X, Huang MN, et al. Low pretreatment prognostic nutritional index predicts unfavorable survival in stage III-IVA squamous cervical cancer undergoing chemoradiotherapy. BMC Cancer. 2025; 25: 377. https://doi.org/10.1186/s12885-025-13752-6.

[38] Yan Q, Wu M, Zhang J, Yang J, Lv G, Qu B, et al. MRI radiomics and nutritional-inflammatory biomarkers: a powerful combination for predicting progression-free survival in cervical cancer patients undergoing concurrent chemoradiotherapy. Cancer Imaging: the Official Publication of the International Cancer Imaging Society. 2024; 24: 144. https://doi.org/10.1186/s40644-024-00789-2.

[39] de Rijke JM, van der Putten HWHM, Lutgens LCHW, Voogd AC, Kruitwagen RFPM, van Dijck JAAM, et al. Age-specific differences in treatment and survival of patients with cervical cancer in the southeast of The Netherlands, 1986-1996. European Journal of Cancer (Oxford, England: 1990). 2002; 38: 2041–2047. https://doi.org/10.1016/s0959-8049(02)00315-5.

[40] Pelkofski E, Stine J, Wages NA, Gehrig PA, Kim KH, Cantrell LA. Cervical Cancer in Women Aged 35 Years and Younger. Clinical Therapeutics. 2016; 38: 459–466. https://doi.org/10.1016/j.clinthera.2016.01.024.

[41] Rutledge FN, Mitchell MF, Munsell M, Bass S, McGuffee V, Atkinson EN. Youth as a prognostic factor in carcinoma of the cervix: a matched analysis. Gynecologic Oncology. 1992; 44: 123–130. https://doi.org/10.1016/0090-8258(92)90027-g.

[42] Yoo J, Choi JY, Moon SH, Bae DS, Park SB, Choe YS, et al. Prognostic significance of volume-based metabolic parameters in uterine cervical cancer determined using 18F-fluorodeoxyglucose positron emission tomography. International Journal of Gynecological Cancer: Official Journal of the International Gynecological Cancer Society. 2012; 22: 1226–1233. https://doi.org/10.1097/IGC.0b013e318260a905.

[43] Akkus Yildirim B, Onal C, Erbay G, Cem Guler O, Karadeli E, Reyhan M, et al. Prognostic values of ADC_mean and SUV_max of the primary tumour in cervical cancer patients treated with definitive chemoradiotherapy. Journal of Obstetrics and Gynaecology: the Journal of the Institute of Obstetrics and Gynaecology. 2019; 39: 224–230. https://doi.org/10.1080/01443615.2018.1492528.

[44] Hong JH, Min KJ, Lee JK, So KA, Jung US, Kim S, et al. Prognostic Value of the Sum of Metabolic Tumor Volume of Primary Tumor and Lymph Nodes Using 18F-FDG PET/CT in Patients With Cervical Cancer. Medicine. 2016; 95: e2992. https://doi.org/10.1097/MD.0000000000002992.

[45] Onal C, Guler OC, Reyhan M, Yapar AF. Long-term outcomes of cervical cancer patients with complete metabolic response after definitive chemoradiotherapy. Journal of Gynecologic Oncology. 2021; 32: e74. https://doi.org/10.3802/jgo.2021.32.e74.