Phase Angle and Nutritional Status: The Impact on Survival and Health-Related Quality of Life in Locally Advanced Uterine Cervical Cancer

The phase angle, an indicator of muscle mass status and membrane cell integrity, has been associated with low survival, poorer clinical outcomes, and worse quality of life among cancer patients, but information on women with uterine cervical cancer (UCCa) is scarce. In this prospective study, we used a bioelectrical impedance analyzer to obtain the PA of 65 women with UCCa. We compared the health-related quality of life and inflammatory and nutritional indicators between low PA and normal PA. The mean age was 52 ± 13. The low PA and normal PA groups differed in terms of the C-reactive protein (15.8 ± 19.6 versus 6.82 ± 5.02, p = 0.022), glucose (125.39 ± 88.19 versus 88.78 ± 23.08, p = 0.021), albumin (3.9 ± 0.39 versus 4.37 ± 0.30, p = 0.000), EORTC QLQ-C30 loss of appetite symptom scale score (33.33 (0.0–100.00) versus 0.0 (0.0–0.0), p = 0.005), and EORTC QLQ-CX24 menopausal symptoms scale score (0.0 (0.0–33.33) versus 0.0 (0.0–100.0), p = 0.03). The main finding of the present study is the interaction between PA and obesity as critical cofactors in the UCCa adeno and adenosquamous histologic variants, to a greater extent than cervical squamous cell carcinoma.


Introduction
Uterine cervical cancer (UCCa) is the second most common cancer in women aged over 65 worldwide. However, in emerging economies, such as Latin America, Asia, and Africa, the women affected are younger, and the impact on these women and their families is more profound [1].
A considerable percentage of newly diagnosed cases in developing countries are detected at a locally advanced clinical stage (IB2 to III, according to FIGO staging) or even a metastatic stage. The clinical conditions described above prevent these women from being candidates for standard surgical treatment according to the NCCN and ESGO criteria in the US and EU. They should be treated by concurrent chemoradiotherapy [1,2].
Limited access to health services and high migration rates are common among women with cervical cancer in developing countries. They have limited treatment options, and this condition profoundly affects survival. In some cases, they undergo hysterectomy, associated with radiosensitizing chemotherapy, radiotherapy, and brachytherapy [3,4].
In clinical settings, the health team assumes that malnutrition in UCCa is a common condition that directly impacts body weight [5]. In the case of UCCa, only a few reports identified in our literature search described phase angle (PA) alterations in women with uterine cancer. Nevertheless, these reports highlighted the role of the phase angle as an anthropometric marker for malnutrition and sarcopenia. The phase angle evaluates the loss of muscle mass and the integrity of the membrane cells. Simultaneously, the reports also emphasized the participation of fat mass as a source of inflammatory mediators and molecules involved in malignant transformation and processes such as migration, proliferation, and metastases [6][7][8][9][10][11].
The gold standard for body composition assessment in clinical settings is magnetic resonance imaging (MRI) or computed tomography (CT). However, one of the most efficient methods is bioelectrical impedance analysis (BIA), which is considered an excellent marker of the function of the cellular membrane, with the additional advantage of inferring its integrity, which is considered a global marker of health [12,13]. All of these techniques allow us to identify a loss of skeletal muscle mass, even in patients with overweight and obesity, associated with the phenomenon of sarcopenic obesity.
Even with the best treatment options, clinical outcomes and patient-reported outcomes can be affected by the patient's nutritional status, including survival rate, functionality, and health-related quality of life (HRQoL), while the causes remain unclear [1,2,13]. However, we do not yet know how profound the impacts of sarcopenic obesity and obesity on UCCa are. There are few literature reports about sarcopenia, sarcopenic obesity, or obesity and their relationships with UCCa adenocarcinoma, in contrast to squamous histologies.
The current report aims to describe the relationship between the nutritional state, phase angle, biochemical markers, functionality, and health-related quality of life in women with UCCa and its histologic type.

Materials and Methods
The Institutional Review Board of the Instituto Mexicano del Seguro Social (Comisión Nacional de Investigación Científica) approved this study (registration number R-2017-785-038). All procedures were carried out in accordance with the principles of the Declaration of Helsinki, and all the patients signed an informed consent form to participate.
We included a prospective cohort of 65 patients with a histological confirmation of locally advanced UCCa, who were treated with radiotherapy because they were not candidates for surgical treatment, and they were followed for three years. All participants were treated at a tertiary hospital in Guadalajara, Mexico. We excluded patients with (1) two or more malignant neoplasms, (2) AIDS/HIV infection, (3) autoimmune disease, (4) any chronic disease, (5) failure to sign the informed consent form, or (6) any contraindication for performing bioelectrical impedance analysis (weight >300 kg, metal prostheses, electronic implants, edema, or limb amputations).
The patients' sociodemographic data, information on the clinical stage and anatomical location of the tumor, and information on the treatment modality were obtained from their medical records.
An experienced dietitian performed the body composition analysis. Measurements of height (m) and weight (kg) were taken with a SECA 213 device (Seca, Hamburg, Germany). A BIA device mBCA SECA 514 (Seca, Hamburg, Germany) was used to measure the phase angle (PA), body mass index (BMI), lean mass percentage, and skeletal muscle mass (SMM).
PA is a BIA-subrogated measure that uses the resistance and reactance generated by body fluids and cell membranes (capacitance) in the human body. PA is understood as the opposition of a structure/tissue to the electrical current through the body and the resistance of the cell membrane. The normal phase angle cut-off was set to 4.45 • .
The laboratory measurements were carried out as part of clinical routine and then abstracted from the medical records. The values of creatinine, C-reactive protein, hemoglobin, absolute lymphocyte count, total cholesterol, total proteins, serum albumin, and globulin were included [14].
We evaluated the health-related quality of life (HRQoL) using validated questionnaires, namely the Mexican-Spanish version of (A) the EORTC (European Organization for Research and Treatment of Cancer) QLQ C-30 (core questionnaire, Fayers PM, 2001) and (B) the EORTC QLQ-CX24 Cervix Uteri-Cancer-specific module [15,16].
The specific module for cervical cancer (EORTC QLQ-CX24) has four multi-item scales to evaluate functional scales (body image, sexual activity, sexual enjoyment, and sexual functionality) and five single-item symptom scales (symptom experience, lymphoedema, peripheral neuropathy, and menopausal and sexual concerns) [17,18]. For both questionnaires, each multi-item scale contains a different set of items, and no item is contained in more than one scale. All the scales are structured similarly.
We followed the instructions in the EORTC QLQ-C30 Evaluation Guide to calculate the final score. The first step was to calculate the mean of the items (that of each item contributing to the construction of the scale) and obtain the raw score. After that, we applied the linear transformation formula to standardize the raw score and transform it into a scoring system ranging from 0 to 100 [16]. A higher score on a functioning scale represents a better quality of life. In contrast, a higher score on the symptom scales indicates severe symptoms/problems and a worse quality of life.
Data were collected using the Excel package, and the statistical analysis was performed using the IBM SPSS Statistics Version 28 (IBM, Armonk, NY, USA) software package. The variables with a parametric distribution were described as means and standard deviations, and differences between groups were calculated by Student's t-test or the ANOVA test. For the variables with nonparametric distributions, we calculated the median and interquartile intervals and used the Mann-Whitney U test and the Kruskal-Wallis test to identify differences between groups. The categorical variables were expressed as proportions and percentages, and the chi-square test or Fisher's exact test was used to compare them between the patients with normal and low PA. Cronbach's alpha value was used to assess the reliability in the case of the multi-item scales of the EORTC questionnaires. All statistical analyses were two-tailed, and a p-value < 0.05 was considered significant.

Clinical and Anthropometric Characteristics
Twenty-nine (44.6%) women had a low PA. We found no differences in the histological and clinical features between the low-PA and normal-PA groups (Table 1). The body composition phenotype analysis showed a higher incidence of sarcopenia (66.7%) and sarcopenic obesity (88.9%) in the low-PA group than in the normal-PA group (p = 0.000). In contrast, women with obesity were more common in the normal-PA group (p = 0.000) ( Table 1).

EORTC QLQ-C30
The functioning scales in the normal-PA group had median scores higher than 80, with the exception of the global health status/quality of life (75.00 (58.33-100.00)) and emotional (66.67 (41.67-91.67)) functioning scales. We found no differences in the functioning scales scores compared to the low-PA group.
Higher scores on the symptom scales were observed for fatigue, pain, loss of appetite, and financial difficulties. The low-PA group had higher scores on the loss of appetite symptom scale than the normal-PA group (p = 0.005).

EORTC QLQ-CX24
The body image and sexual/vaginal functioning symptom scales showed higher scores in the EORTC QLQ-CX24 questionnaire. The scores on the menopausal symptom scale were higher in the normal-PA group than in the low-PA group (p = 0.039) (see Table 3).

Survival Status among the Cohort of Cervix-Uteri Cancer Patients Survival
The survival of the cohort of women with cancer showed no significant differences according to the PA group (Table 4).

Discussion
In cancer patients, body composition analysis has been proven to be a better indicator of nutritional status than BMI. BIA provides a tool for measuring the body composition through a reproducible technique that can be performed quickly and easily at the bedside. It was used by our research team for patients with other types of cancer and has been proven to be effective among these populations [8,11].
PA, a BIA-subrogated measure, represents the resistance created by body fluids and cell membranes (capacitance) in the human body. It is correlated positively with capacitance and negatively with resistance [2], being considered an excellent marker of the cell membrane function. A low PA is associated with decreased cell integrity or cell death/apoptosis, while a high PA is associated with an intact cell and better cell membrane function [7,8,17].
The evidence presented by the cohort investigated here shows the associations of the phase angle and nutritional status with survival, the quality of life, body composition phenotype, biochemical biomarkers, and histological characteristics.
In our previous studies, we used the ROC curve tool to estimate the cutoff for PA. We used the same layout for the women who were evaluated in the present study. The normal phase angle cut-off was set to 4.45, which coincides with the Youden index of the ROC curve (cut-off point with the maximum Kolmogorov-Smirnoff), corresponding to a PA of 4.45 and an overall model quality score of 0.55 (sensitivity 0.600; specificity 0.533).
The PA values of healthy groups can vary between different racial groups (higher values in African and Caucasian groups and lower values in Asian and Latino populations), genders (higher values in men than in women), and ages (lower values after 60 years) [8,19].
Healthy people have a PA between 4 and 10 degrees, but this differs according to sex, age, weight, BMI, and fat percentage categories. These reference values can provide a basis for phase angle evaluations in the clinical setting [8,16,20,21].
The PA values of a Mexican study population were evaluated based on healthy women aged 18-82 years with a BMI between 18-31 kg/m 2 . The PA had a mean of 6.36 • ± 0.97 • [13,22]. Our current results yielded data consistent with our previous findings and the findings of other Mexican research groups, who found that Mexican women have a lower PA than other demographics, such as European, Afro-descended, and even Asian women [7,23].
The evaluation of PA aids in the follow-up of oncological chronic diseases, and it is a crucial tool for predicting the outcomes, frailty, functionality, adverse prognosis, response to treatment, and survival/mortality of UCCa women [19].
A secondary care hospital in San Luis Potosi, Mexico, evaluated 70 women in an observational cross-sectional study to assess the PA and body composition of women with a diagnosis of UCCa using electrical vector bioimpedance. The results showed a PA mean of 4.66 ± 0.87 • (with a range of 2.9 • to 6.2 • ). In our results, a low PA score was correlated with worse clinical stages (III and IV n = 20; 69%) and worse body composition phenotype (sarcopenia and sarcopenic obesity, see Table 1) [7,8].
Risk cofactors may explain the higher prevalence of obesity-associated low PA in our UCCa patients, such as smoking, malnutrition, metabolic disorders, and a sedentary lifestyle [8,11,20,24,25]. All these cofactors are associated with an impaired immune response and inflammatory mediators, which were significantly altered in our studied population (see Table 2, C-reactive protein values in women with PA < 4.45: 15.84 mg/L versus 6.82, observed in women with PA over the cutoff; p = 0.022).
Another mechanism identified was the percentage of adipose tissue. The number of adipose cells is related to premalignant and malignant tissue transformation in obese women. Chronic systemic inflammation due to obesity induces cell membrane damage and is also a metabolic source of inflammatory mediators (see Table 2) [8,9,17,23,26,27].
A meta-analysis published in 2016 demonstrated the link between cervical cancer and obesity. Based on the findings, obese women have a slightly increased risk of developing uterine cervical carcinoma [28]. In our current study, we identified obesity in 38.4% of the women with UCCa and in 27.7% of those with sarcopenic obesity. It is important to emphasize that the phenomenon of sarcopenic obesity is associated with a PA of <4.45 (see Table 1).
In these conditions, it is important to highlight the increased likelihood that in an ethnic or geographical group with a high prevalence of overweight-obesity (female rates in Mexico reach over 70%), sarcopenia may be associated with obesity (sarcopenic obesity), which aggravates the inflammatory conditions and causes a loss of functionality and quality of life (see Table 3) and worse clinical outcomes (see Table 2).
In UCCa, the histological type of adenocarcinoma usually responds poorly to treatment compared to squamous cell carcinoma. In our population, we observed a relationship between the body composition phenotype and histological types. Obesity was more frequently associated with adenocarcinoma and adenosquamous cervical carcinoma (14/34 (41.2%)) than squamous cell carcinoma, and sarcopenic obesity was associated with 22.2% of cases of adenocarcinoma and adenosquamous cervix-uteri carcinoma.
The biological behavior of UCCa in this group of women with obesity and sarcopenic obesity resembled the observed characteristics of endometrial cancer with diabetes and insulin resistance related to obesity.
Future studies of women with UCCa will be crucial in order to determine the interaction of the genotype of HPV infection with the overweight-obesity phenomenon, compared to the absence of HPV infection in the different histological variants of cervical cancer, and to rule out endometrial cancer with cervical invasion [20,29,30].
Carriers and survivors of UCCa show a reduction in their physical autonomy in terms of functionality due to a loss of muscle mass and the presence of symptoms related to the pathology and its treatment, and this situation directly affects their quality of life [16,18]. Based on the observed differences in the scores on the functional and symptom scales of the EORTC QLQ-C30 questionnaire among women with cervical cancer, we can hypothesize that malnutrition, coupled with cancer progression, surgery, and adjuvant therapy, together with their side effects, profoundly impairs quality of life [31].
In women with previously deteriorated HRQoL due to overweight or obesity and a loss of muscle mass, PA could be a clinical predictor of prognosis. Based on this, early nutritional counseling and support to guide physical therapy and exercise programs for overweight and obese patients could help to improve the HRQoL and functionality of patients with UCCa [31,32].

Conclusions
The main finding of the present study is the interaction between PA and obesity as critical cofactors in the UCCa adeno and adenosquamous histologic variants, to a greater extent than cervical squamous cell carcinoma.
The current results offer a basis for evaluating the interaction of the body composition with the HPV infection genotype as a risk factor for the development of cervical adeno and adenosquamous variants of UCCa in a way that resembles endometrial cancer. We conclude that this cohort is a starting point for the assessment of the consistency of obesity as a factor in cervical adenocarcinoma. Funding: This research received no external funding. The study was funded through the funds of the hospital for each of the participating institutions, based on the budget allocated for that purpose, and through the inter-institutional collaboration between the IMSS and the Universidad de Guadalajara.

Institutional Review Board Statement:
This study was carried out according to the Declaration of Helsinki guidelines and was approved by the Institutional Review Board of the Comité Local de Investigación en Salud 1301, Instituto Mexicano del Seguro Social, México, through the project: "Calidad de vida relacionada a la salud en pacientes con cáncer cervico uterino atendidas en el IMSS", con número de resgistro R-2017-785-038.

Informed Consent Statement:
This study was carried out according to the Declaration of Helsinki guidelines. It was approved by the Institutional Review Board of the Comite Local de Investigación en Salud 1301, Instituto Mexicano del Seguro Social, Mexico, R-2017-785-038.

Data Availability Statement:
The datasets generated and analyzed in the current study are not publicly available because they are the property of the Instituto Mexicano del Seguro Social. Institutional and federal bodies restrict unlimited access to personal data, but they are available from the corresponding authors on reasonable request with prior authorization from the institution.