Article Text
PDF
Abstract
Objective Cancer patients are at risk of malnutrition, which is associated with poor oncological outcomes. The aim of this study was to assess the incidence of malnutrition before, during, and after radiotherapy in locally advanced cervical cancer patients. In addition, we evaluated the impact of malnutrition on survival, and whether and when malnourished patients were referred to a dietitian.
Methods This retrospective cohort study included cervical cancer patients who received primary or adjuvant radiotherapy with curative intent between January 2013 and January 2021. Patient and treatment characteristics, including longitudinal data on weight and dietary care, were retrieved from the electronic patient files. Malnutrition was defined by body mass index and weight loss according to the Global Leadership Initiative on Malnutrition (GLIM). Overall survival was estimated using the Kaplan–Meier method. Cox proportional hazard regression analysis was used to estimate hazard ratios for key prognostic factors.
Results A total of 294 patients were included. Median follow-up was 40 months (range 7–101 months). Malnutrition occurred in 44 patients (15%) at baseline, in 132 (45%) during radiotherapy, and in 63 (21%) during follow-up. Referral to a dietician occurred in 45% of the 138 patients who were malnourished before or during radiotherapy. Malnutrition was significantly associated with worse survival after adjusting for age, performance score, diabetes, histology, International Federation of Gynecology and Obstetrics (FIGO) stage, and nodal stage. The 3 year overall survival in patients with malnutrition was 77% (95% confidence interval (CI) 70% to 85%) and without malnutrition 89% (95% CI 83% to 95%); p=0.001). Independent significant risk factors for worse overall survival were: malnutrition, age ˃52 years, adenocarcinoma, FIGO stage III/IV, and N1 disease.
Conclusion Malnutrition was common in cervical cancer patients treated with radiotherapy and was associated with a shorter overall survival. Further studies are needed to evaluate the effectiveness of better monitoring of malnutrition and faster and better dietary intervention on survival and quality of life.
- Uterine Cervical Neoplasms
- Radiotherapy
- Radiation Oncology
Data availability statement
Data are available upon reasonable request. Research data can be made available on request to the corresponding author, but are not currently freely available online.
Statistics from Altmetric.com
WHAT IS ALREADY KNOWN ON THIS TOPIC
Malnutrition is associated with poorer survival in gynecological cancer patients treated with chemotherapy and surgery.
Few studies have evaluated malnutrition during radiotherapy and its effect on survival in cervical cancer patients.
WHAT THIS STUDY ADDS
Malnutrition in cervical cancer patients treated with radiotherapy was associated with worse overall survival even after adjusting for several potentially confounding factors.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY
Future studies should evaluate whether early detection of malnutrition and timely intervention improves the prognosis of cervical cancer patients.
Introduction
Each year, approximately 900 women in The Netherlands are diagnosed with cervical cancer.1 Patients with locally advanced disease receive radiotherapy combined with weekly cisplatin based chemotherapy. Approximately 80% of patients experience diarrhea and anorexia as common side effects of chemoradiation.2 These symptoms, combined with malignancy induced metabolic dysregulation, place patients at risk of developing malnutrition.
In cervical cancer patients, a low body mass index before radiotherapy is associated with an increased risk of developing severe gastrointestinal symptoms.3 4 For every 1 point decrease in body mass index, this risk increases by 7%.4 This may be explained by the greater amount of intra-abdominal fat in women with a higher body mass index, resulting in less bowel volume near the irradiated target volume.5 In addition, irradiation of a larger bowel volume is associated with an increase in severe acute gastrointestinal symptoms such as diarrhea. This, in turn, can lead to weight loss and poor nutritional status, creating a downward spiral.5
Malnutrition is a common consequence of cancer and cancer treatment. Some studies estimate that at least half of all cancer patients are malnourished.6 The incidence and risks of malnutrition have been studied mainly in cancer patients undergoing surgery or systemic therapy. Perioperative weight loss and poor nutritional status are correlated with both worse survival and increased toxicity in patients with bladder and gastric cancer.7 8 Weight loss during chemotherapy is also associated with worse overall survival in ovarian cancer patients.9
Only a few small studies, each with <110 patients, have described malnutrition and weight loss in this specific patient population.10–12 The aim of the present study was to evaluate weight loss and malnutrition before, during, and after radiotherapy in a large cervical cancer patient cohort. We also assessed the impact of malnutrition on survival and whether and when patients who were malnourished were referred to a dietitian.
Methods
This retrospective cohort study consecutively included patients with histologically proven cervical cancer who were treated with curative intent with primary or adjuvant external beam radiotherapy between January 2013 and January 2021. Other inclusion criteria were a minimum follow-up of 3 months after completion of radiotherapy, and data available for body weight at baseline, during radiotherapy, and at follow-up.
Treatment consisted of external beam radiotherapy in 23–28 fractions of 1.8–2.0 Gy, generally combined with weekly cisplatin 40 mg/m2 or once weekly with deep hyperthermia if contraindicated, followed by a brachytherapy boost. In a primary setting, the external beam radiotherapy field covers the uterus, parametria, upper vagina, and the pelvic lymph node region. The external beam radiotherapy field was extended to the para-aortic region up to the renal veins in the case of pathological lymph nodes in the common iliac or para-aortic region. Elective groin irradiation was indicated if the lower third of the vagina was involved. Patients with pathologically enlarged lymph nodes received an external beam radiotherapy boost up to a dose of 60 Gy.
Adjuvant radiotherapy after oncological surgery (usually radical hysterectomy Querleu–Morrow type C) was indicated in the case of pathological lymph nodes, positive surgical margins, and parametrial invasion. In the adjuvant setting, the clinical target volume covers the tumor bed, including the upper vagina, and the draining pelvic lymph nodes. Patients were clinically evaluated by gynecological examination every 3 months in the first year, every 6 months in the second and third years, and annually thereafter until 5 years after treatment.
Malnutrition during radiotherapy or follow-up was based on both body mass index and weight loss, as defined by the Global Leadership Initiative on Malnutrition (GLIM) criteria.13 All patients <70 years of age at diagnosis with a body mass index <20 kg/m2, and ≥70 years of age with a body mass index <22 kg/m2, were classified as malnourished. In addition, patients with >5% weight loss at the end of radiotherapy (compared with baseline) or >10% weight loss during the first year of follow-up (compared with baseline) were also classified as malnourished. If patients relapsed during follow-up, they were censored for malnutrition at the onset of symptoms that ultimately led to a relapse.
For this analysis, weight was recorded at three time points: (1) at baseline, (2) during radiotherapy, and (3) during follow-up. Baseline weight was defined as the weight reported at the first consultation with the gynecologist or radiation oncologist. The weight recorded in the patient file closest to the last fraction of radiotherapy (between 2 weeks before and 1 week after) was selected as the weight during radiotherapy. The weight recorded closest to 12 months after radiotherapy was selected as the weight during follow-up. Malnutrition at baseline was assessed using body mass index only.
Patient and treatment characteristics were obtained from patient records. All types of surgery associated with a risk of intra-abdominal adhesions were included as major abdominal surgery.4 As a proxy for socioeconomic status, postal code was collected and correlated with status scores for that area (based on income, education, and unemployment rates) supplied by The Netherlands Institute for Social Research.14 The mean status score of The Netherlands was used as a cut-off between high and low socioeconomic status. Overall treatment time was defined as the time in days between the first up to the last fraction of either external beam radiotherapy or brachytherapy.
Univariable and multivariable Cox proportional hazards regression analysis was used to assess the association between malnutrition and overall survival. Missing values were handled by pairwise deletion for univariable analysis and listwise deletion for multivariable analysis. This resulted in a complete case analysis. Backward conditional multivariable analysis was performed for variables below the univariable p value of 0.10. Variable elimination was assessed using both Wald and log likelihood ratios. Proportional hazards assumption was assessed graphically by evaluating log minus log hazard function plots.
The Kaplan–Meier method was used for analysis of overall survival, and the log rank test for analysis of differences between groups.15 Survival was calculated as months from diagnosis (pathology date) to death or the last follow-up. A prespecified sensitivity analysis was performed to examine the effect of malnutrition on cancer specific survival. To adjust for the baseline imbalance in International Federation of Gynecology and Obstetrics (FIGO) stage between patients with and without malnutrition, a post hoc sensitivity analysis was performed. Using the Kaplan–Meier method, we analyzed the effect of malnutrition (excluding baseline malnutrition since this could hypothetically result from high tumor burden) on overall survival separately in FIGO stage I/II and FIGO stage III/IV patients. A post hoc sensitivity analysis was also performed to assess the effect on overall survival of the two definitions of malnutrition (low body mass index and weight loss) separately. Data were analyzed using SPSS V.26.0 (IBM Corp. Released 2019. Armonk, New York, USA) and R V.4.1.1.16
Results
Patient and Treatment Characteristics
Between January 2013 and January 2021, 404 cervical cancer patients were referred to our center for curative radiotherapy, of whom 294 were available for analysis (online supplemental Figure 1). Table 1 shows patient and treatment characteristics. Median follow-up was 40 months (range 7–101 months). In general, patient and treatment characteristics were well balanced between the malnourished and never malnourished patients. Patients with FIGO stage III and those who had been treated with hyperthermia were more often malnourished (Table 1).
Supplemental material
Supplemental material
Baseline patient and treatment characteristics in patients with and without malnutrition before, during, or after radiotherapy
Malnutrition, Weight, and Dietetic Referral
The incidence of malnutrition and referral to the dietitian at baseline, during radiotherapy, and during follow-up is shown in Table 2. In 9 of the 45 patients treated with adjuvant radiotherapy, baseline body mass index was calculated based on their weight before surgery. Fourteen patients developed a recurrence during follow-up and were censored for malnutrition. A total of 138 patients (47% of 294) had malnutrition at baseline or during radiotherapy. Malnutrition occurred at some time before, during, or after treatment in 50% of all patients (148/294, one patient was censored). Figure 1 shows the changes in the presence of malnutrition per patient over time. Of the 294 patients, 189 (64%) lost weight during radiotherapy, with a median weight loss of 1.9 kg. Weight loss during follow-up (compared with baseline) occurred in 114 of the 294 patients (49%).
Alluvial diagram showing the longitudinal changes in the occurrence of malnutrition in the study cohort. The x axis represents all 294 patients included in the cohort who could be followed longitudinally over time (from left to right) along the x axis. The y axis represents the number of patients. The stream fields between time points represent the changes in diagnosis of malnutrition over time—for example, the white dotted line represents a patient that was malnourished (red) both before (baseline) and during radiotherapy, but was not malnourished (blue) during follow-up.
Number of malnourished patients and referral to dietitian at different time points before, during, and after radiotherapy (n=294)
Of the 148 patients who were malnourished at any time, 96 (65%) were referred to a dietitian at some point. Out of the 138 patients who were malnourished at baseline or during radiotherapy, 45% (62 patients) were referred to a dietitian. The first consultation took place on the same day in 33% of cases (32/96). Nine patients (9%) did not receive dietary counseling within 2 weeks of referring. Two patients failed to see a dietitian despite being referred.
Survival
In the overall cohort of 294 patients, overall survival at 3 and 5 years was 83% and 72%, respectively. The 5 year overall survival was 85%, 79%, 60%, and 36% in patients with FIGO stages I, II, III, and IV, respectively. In patients with malnutrition (either at baseline, during radiotherapy, or during follow-up), the 3 year and 5 year overall survival rates were 77% (95% confidence interval (CI) 70% to 85%) and 60% (95% CI 49% to 73%), respectively. In patients without malnutrition, overall survival was 89% (95% CI 83% to 95%) and 84% (95% CI 49% to 73%), respectively (p=0.001) (Figure 2).
Kaplan–Meier curve showing overall survival for patients with (red) and without (blue) malnutrition at baseline and/or during radiotherapy and/or follow-up. The shaded areas represent the 95% confidence intervals.
In the 138 patients with malnutrition at baseline and/or during radiotherapy, mean overall survival after referral to a dietitian was 66 months compared with 63 months without referral (p=0.4). In the sensitivity analysis, there were no significant differences in overall survival (p=0.063 in FIGO stage I/II and p=0.140 in FIGO stage III/IV) (online supplemental Figure 2). Online supplemental Figure 3 shows that both malnutrition based on body mass index and malnutrition based on weight loss were significantly associated with poor overall survival (p=0.003 and p<0.001, respectively). Cancer specific survival was not significantly worse in malnourished patients (online supplemental Figure 4) (p=0.06).
Supplemental material
Supplemental material
Supplemental material
Table 3 shows the univariate and multivariate Cox regression analysis for overall survival. Age ˃52 years, adenocarcinoma (vs squamous cell carcinoma), FIGO III/IV (vs FIGO I/II), N1 (vs N0), and malnutrition were independent significant risk factors for worse overall survival.
Univariable and multivariable Cox regression analysis for overall survival (n=294)
Discussion
Summary of Main Results
Malnutrition in women treated with curative intent for locally advanced cervical cancer increased to 45% during radiotherapy. Although survival was significantly worse in women with malnutrition, with a 12% absolute decrease in 3 year overall survival, less than half of these women received appropriate and timely referral for dietary care.
Results in the Context of Published Literature
To date, there is only one study that has evaluated weight loss during radiotherapy and its effect on the survival of patients with cervical cancer.10 Of the 108 patients included, 14% lost weight (defined as >10% loss of body weight). They also found significantly worse overall survival in patients with weight loss during treatment, with a hazard ratio (HR) of 2.37 (HR 2.17 in our cohort). Only 9% compared with 65% in our cohort were referred for dietary care. Note that this study was conducted in the US and patients had a higher body mass index at baseline (mean 29 vs median 26 body mass index points in our cohort). Furthermore, malnutrition was only assessed by weight loss and only during treatment. Clark et al showed that low body mass index was an important poor prognostic factor in cervical cancer patients in a large cross sectional study. Patients with a pretreatment body mass index <18.5 had significantly worse overall survival.17 In the only prospective study evaluating weight and body composition during radiotherapy in cervical cancer patients (n=49), underweight was significantly more common at the end of treatment (11%) than at baseline (0%). They found no significant changes in body composition before and after treatment.18 The effect on survival was not evaluated in this study due to the small sample size. In a retrospective study by Ohno et al, weight was assessed weekly during radiotherapy in 73 cervical cancer patients.12 They found that weight loss occurred throughout treatment, as early as in the first week, simultaneously with symptoms of anorexia and diarrhea.
Strengths and Weaknesses
By including almost three times as many patients as in previous studies, we were able to control for a number of confounding factors through multivariable analysis. Previous studies investigating the relationship between malnutrition and survival generally used cross sectional measurements of weight or body mass index (usually at baseline) and did not consider weight loss during and after treatment.3 17 The studies that have been published to date that evaluated weight during the course of radiotherapy in patients with cervical cancer all described a non-Western European population.10–12
The relationship between malnutrition and poor oncological outcome is complex, as malnutrition is also correlated with other prognostic factors, such as lifestyle, performance status, comorbidities, tumor metabolism, and tumor stage.4 19–21 However, in the multivariable analysis, malnutrition remained an independent and strong risk factor for worse overall survival, even after adjusting for several factors, including tumor stage. In our study, malnourished patients more frequently received hyperthermia, probably reflecting older age and higher incidence of (renal) comorbidities. To assess the robustness of our findings, we performed several sensitivity analyses. Although not statistically significantly different, cancer specific survival seemed to be worse in malnourished patients. Overall survival was consistently worse in malnourished patients independent of FIGO stage. The baseline weight of nine patients (3% of 294) was recorded before surgery. Therefore, any weight loss in these patients may have occurred perioperatively.
Due to the retrospective design of the study, some data on weight were missing. This is an important finding as it shows that weight loss is often underappreciated as a risk factor. This is also reflected in the fact that not all patients were referred for dietary advice when significant weight loss was recorded.22 Weight loss before diagnosis can only be quantified on the basis of the medical history and is therefore dependent on the accuracy of the patient’s memory. With these limitations, we still found that 15% of patients were underweight at baseline, more than seven times the incidence in the Dutch reference population.23
Implications for Practice and Future Research
This study alone does not prove the effectiveness of better monitoring for malnutrition and intervening earlier and better. However, it does lend support to the idea that an appropriate and timely referral to a dietitian can lead to an improved oncological outcome. The effectiveness of dietary interventions on quality of life and prognosis should be the subject of further research.
Conclusions
This study of women undergoing radiotherapy for locally advanced cervical cancer confirms our hypothesis that malnutrition is common in these patient and is a significant independent risk factor for poor survival. Unfortunately, weight was not always recorded consistently and referral to a dietitian was low among women who were malnourished.
Data availability statement
Data are available upon reasonable request. Research data can be made available on request to the corresponding author, but are not currently freely available online.
Ethics statements
Patient consent for publication
Ethics approval
This study involves human participants. The Medical Ethics Committee of Amsterdam UMC location AMC/University of Amsterdam exempted this study. In the present study, participants were not subjected to any procedures or required to follow any rules of behavior. In addition, the data were collected years after diagnosis and for a large number of patients it was not feasible to obtain informed consent retrospectively.
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Footnotes
Contributors Conceptualization: JL, LvL, LJAS, and HW. Data curation: JL. Formal analysis JL. Investigation: JL. Methodology: JL, BP, and LJAS. Project administration JL. Resources: LvL, KH, BP, LJS, and HW. Supervision: LvL, LJS, and HW. Validation: LvL and ID. Visualization: JL. Writing—original draft: JL. Writing—review and editing: LvL, ID, KH, BP, LJAS, and HW. Guarantor HW. All authors have read and agreed to the published version of the manuscript.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial, or not-for-profit sectors.
Competing interests BP: grant ELEKTA (payment to the institution) and payment by BD (to the institution).
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.