Impact of Chemotherapy Regimens on Body Composition of Breast Cancer Women: A Multicenter Study across Four Brazilian Regions

This study aimed to investigate the effect of chemotherapy (CT) and its different types of regimens on the anthropometry and body composition of women with breast cancer. Three-hundred-and-four women with breast cancer were enrolled in this multicenter study. The participants were evaluated before the infusion of the first cycle of CT (pre-CT), and until two weeks after CT completion (post-CT), regarding body weight, body mass index (BMI); waist circumference (WC); waist-to-height ratio (WHtR); conicity index (C-index); fat mass index (FMI); and fat-free mass index (FFMI). CT regimens were classified as anthracycline-based (AC—doxorubicin or epirubicin); anthracyclines and taxane (ACT); cyclophosphamide, methotrexate, and 5-fluorouracil (CMF); or isolated taxanes (paclitaxel or docetaxel). Women significantly increased BMI and FMI post-CT (p < 0.001 and p = 0.007, respectively). The ACT regimen increased FMI (p < 0.001), while FFMI increased after AC (p = 0.007). It is concluded that the CT negatively impacted body composition and the type of regime had a strong influence. The ACT regimen promoted an increase in FMI compared to other regimens, and the AC increased FFMI. These findings reinforce the importance of nutritional monitoring of breast cancer patients throughout the entire CT treatment.


Introduction
Brazil presents a multidimensional socioeconomic context throughout its five regions. Socioeconomic status (SES)-comprised of factors such as income, education, poverty, and wealth-can be different through the regions; however, the overall combination of these factors represents the whole Brazilian nation. The SES can be linked to the prevalence in rates of and mortality from chronic diseases [1]. For instance, breast cancer is the most University of Pelota (RS); and the Carmela Dutra Maternity Hospital and Oncology Research Centre (SC). Each center had its own research protocol and used a standardized protocol for the recruitment of participants, collection of questionnaire data, and anthropometric and body composition measurements. As the study aims were similar in all centers, the research teams have harmonized the data for this study proposal. The inclusion criteria were women with primary breast cancer undergoing CT without previous treatment for breast cancer or other cancer. All women aged 18 years or over, with clinical stages I to III, presenting available information related to CT treatment, tumor characteristics, anthropometric, and body composition data were included. Exclusion criteria were patients with mobility difficulties and any cognitive or psychiatric impairment that prevented understanding and data collection were also excluded.

Ethics Statement
This study was approved by the Research Ethics Committee from the Clinical Hospital of the Federal University of Goiás (n.3.858.331/2020). Written informed consent was provided by all participants after all risks, discomforts, and benefits involved in the study were reviewed in each region.

Data Collection
Data were collected from 2004 to 2018 following the study protocol in each region. The follow-up time was based on the CT regimen, presenting a range from four to six months. Trained researchers conducted a personal interview at the diagnosis or before the infusion of the first cycle (pre-CT-T0) and after the last CT cycle until two weeks after CT completion (post-CT-T1).

Socioeconomic Status and Tumor-Related Characteristics
Age (full years), self-reported skin color, marital status (with or without a partner), education level attained, family income/month, and menopausal status were collected during a personal interview in T0.
Morphological tumor types (ductal, lobular, mucinous); degree of differentiation, clinical stage (tumor node and metastasis-TNM); molecular subtypes (luminal, human epidermal growth factor receptor (HER2), and triple-negative); type of surgery (mastectomy/ breast-conserving surgery); and CT regimen were obtained from the physician in charge or from medical records. CT regimens were classified as anthracycline-based; combined cyclophosphamide with doxorubicin or epirubicin (AC); anthracyclines plus taxanes (ACT); cyclophosphamide, methotrexate, fluorouracil (CMF); and isolated taxanes (PD-either paclitaxel or docetaxel). The inclusion criteria were women with primary breast cancer undergoing CT without previous treatment for breast cancer or other cancer. All women aged 18 years or over, with clinical stages I to III, presenting available information related to CT treatment, tumor characteristics, anthropometric, and body composition data were included. Exclusion criteria were patients with mobility difficulties and any cognitive or psychiatric impairment that prevented understanding and data collection were also excluded.

Ethics Statement
This study was approved by the Research Ethics Committee from the Clinical Hospital of the Federal University of Goiás (n.3.858.331/2020). Written informed consent was provided by all participants after all risks, discomforts, and benefits involved in the study were reviewed in each region.

Data Collection
Data were collected from 2004 to 2018 following the study protocol in each region. The follow-up time was based on the CT regimen, presenting a range from four to six months. Trained researchers conducted a personal interview at the diagnosis or before the infusion of the first cycle (pre-CT-T0) and after the last CT cycle until two weeks after CT completion (post-CT-T1).

Socioeconomic Status and Tumor-Related Characteristics
Age (full years), self-reported skin color, marital status (with or without a partner), education level attained, family income/month, and menopausal status were collected during a personal interview in T0.

Behavioral Variables
The consumption of alcoholic beverages was calculated in grams per day according to frequency, quantity, and type of drink mentioned as habitual [22], and classified into ≥10 (risk for breast cancer) or <10 g/day alcohol intake (non-risk) [23]. Smoking status was determined based on current smokers (current smokers or quit smoking less than a year ago) or ex-smokers (stopped smoking more than 1 year ago) and non-smokers, based on a previous study [22]. Physical activity (PA) was assessed using the International self-administered Physical Activity Questionnaire short form (IPAQ-SF) [24], and volunteers were classified as active (≥150 min of moderate PA/week) or inactive (<150 min of moderate PA/week) [25]. These dates were collected in T0 for characterizing the sample.

Anthropometry and Body Composition
Anthropometric measurements were conducted according to Habicht procedures [26] and collected at T0 and T1. Height was evaluated by a vertical stadiometer with an accuracy of 0.1 cm (Model P150-C, Lider R, São Paulo, Brazil) and body weight was assessed by a digital scale accurate to 0.1 kg and with a capacity of 150 kg (Lider R or Filizola™, São Paulo, Brazil). BMI was calculated as a ratio between weight (kg) and height (m) 2 using age-specific cut-off values for adults (normal weight: ≥18.5-24.9 kg/m 2 and overweight: ≥25 kg/m 2 ) [27]; and older adults (normal weight: ≥22-27 kg/m 2 and overweight: >27 kg/m 2 ) [28].
Waist circumference (WC) was measured at the midpoint between the lowest rib and the iliac crest using an inelastic measuring tape of 1 mm precision, and women were classified as having a low (<80 cm), high (≥80 cm), and very high (≥88 cm) risk for metabolic complications, according to the World Health Organization [27]. The waist-toheight ratio (WHtR) was calculated as a ratio between WC (cm) and height (cm), and the values ≥0.5 were considered as an indicator of fatness [29]. The conicity index (C-index), an index based on the similarity between accumulated fat around the waist and a cone that rises the metabolic risk, was obtained following the equation proposed by Valdez et al. [30]: . Fat-free mass and fat mass (kg and percentage) were assessed by dual-energy X-ray absorptiometry (DXA; GE© Lunar densitometer, DPX NTVR, with ENCORE 2011 software, version 13.60, GE Healthcare, Chicago, IL, USA) in Goiania; by a tetrapolar single-frequency bioelectrical impedance analyzer (BIA) Biodynamic ® Model 450 (TBW, São Paulo, Brazil) in São Paulo and Fortaleza; and by BIA Quantum (RJL Systems™, Clinton Twp, MI, USA) in Pelotas. FMI and FFMI were calculated as fat mass (kg) and fat-free mass (kg) divided by height squared, respectively [31]. Although different BIA technologies with different algorithms were used to estimate body compartments, only the difference between the two assessments (pre-CT and post-CT) was considered for statistical analysis.

Statistical Analysis
The sample size was calculated based on the article published by Godinho-Mota et. al. [8], which has a similarity with the study aims. For the sample size, G*Power software version 3.1.9.2 was used, taking into consideration the CT effect on total body fat percentage [12]. The effect size of 0.575 showed that, with a significance level of 95% and statistical power of 80% (power 1-β 0.80), the minimum number of participants required was 42 patients. Thus, each region was included if it had at least 42 patients.
Data were processed using Excel software 10.0 (version 2013, Microsoft Corporation, Redmond, WA, USA), and statistical analysis was performed using SPSS software version 21.0 (IBM Corp., Chicago, IL, USA). Descriptive statistics were conducted to characterize the sample studied. Data presented in mean, standard deviation, and frequencies The General Mixed Model (GMM) adjusted by age was used to verify the effect of CT regimens (pre-CT × post-CT) and the interaction between regimen and time on the variables investigated. Estimated marginal means and 95% confidence intervals (CI) were compared in pairs using Sidak for multiple tests. The level of significance for all analyses was set at p < 0.05.

Results
A total of 304 women with a mean age of 51 ± 11 years were evaluated in this study. Most women live with a partner (61.5%), receive from one to six minimum wages (42.8%), and are postmenopausal (40.1%) ( Table 1). Most of the participants had invasive ductal carcinoma (71.4%), stage II (57.9%), luminal subtype (37.8%), and received the ACT regimen (48.7%) ( Table 2).  Most women are physically inactive (34.9%), do not report alcohol consumption (46.7%), and are non-smokers (54.9%) ( Table 3).  Although the anthropometric measurements of central obesity represented by WC, WHtR, and C-index do not present significative changes, a significant increase in BMI (∆ = 0.50 ± 0.09; p < 0.001) and FMI (∆ = 0.36 ± 0.10; p = 0.007) can be observed post-CT, independently of the regimen used (Table 4). Overall, CT regimens have a significant increase in FMI and FFMI. Through the posthoc Sidak test, the effects of the interaction between time and regimen can be found. The results indicate that, for the increase in BMI after CT, time seems to be the factor that has the most influence on this change (p < 0.001); for the increase in FMI after CT, both time (p = 0.007) and type of regime (p = 0.040) seem to influence this parameter of body composition, presenting significance for the ACT regimen (∆ = 0.56 ± 0.15). Finally, an increase in FFMI is probably more related to the type of CT regimen employed than the duration of the treatment itself, presenting significance for the AC regimen (∆ = 0.34 ± 0.12; p < 0.001) ( Table 4).

Discussion
This study found that CT has an effect on BMI and FMI, as both increased after treatment. Regarding CT regimes, FMI increased significantly in women after ACT treatment, while FFMI increased after AC. These findings confirm our initial hypothesis that anthropometric and body composition parameters change throughout CT, and that the impact is different according to the type of CT regime.
Patients evaluated in this study were mainly classified as overweight in all centers. In the actual Brazilian epidemiologic scenario, the proportion of overweight women (+18 years) is ranged from 51.3% to 62.7% [32], a status that is positively associated with breast cancer risk [33,34]. Additionally, Brazilian breast cancer women presented higher BMI post-CT. The literature suggests that weight gain post treatment exists [8,9]; however, a recent cohort of Denmark women did not find an association between weight gained pre and post chemotherapy, even finding an average increase in body weight of 1.2 kg (p = 0.29) [35]. Although no significance was found in relation to the regimen used in our study, a meta-analysis [10] showed an increase of 2.7 kg (95% CI, 2.0-7.5) during CT, especially in women treated with CMF, where the weight gain was 3.5 kg (95% CI, 2.7-4.3) versus 1.4 kg (95% CI, 0.7-2.0) for non-CMF. Evidence suggests that CT regimens, mainly CMF, might conduct in greater body weight gain (8-10 kg), visceral adiposity, and reduction in FFMI regardless of age, energy food consumption, and clinical staging [10,13]. These changes can be driven by hormonal changes, given that estrogen suppression and insulin resistance induction can increase total fat and decrease fat-free mass, including skeletal muscle [36][37][38].
The central adiposity did not differ post-CT in this study. We used WC and WHtR measurements, which are highly used in population-based studies due to the viability of evaluation; but they can be imprecise indicators of intra-abdominal adipose tissue, as well subcutaneous fat deposition and visceral adipose tissue [39]. Computed tomography and magnetic resonance imaging are more accurate methods for assessing abdominal fat in breast cancer patients [40]; however, those methods have a higher cost and are not routinely used to assess breast cancer patients' body composition in Brazil.
In this current study, FMI increased in women post-CT with ACT regimen. The literature shows different findings related to CT regimens used by breast cancer patients on body composition [8][9][10]; however, ACT and CMF regimens are normally longer and can impact body composition [41]. Some women had a fat mass gain, possibly due to the abrupt hormonal changes and the ovarian collapse induced by the CT, which may increase fat mass in the visceral region and reduce bone mineral density [41].
It can be noticed that CT Impacts woman's life independently of the regimen used (AC, ACT, CMF), affecting BMI and FMI values. In addition, CT can promote changes in physical activity and basal metabolic rate, which may be linked to changes in body composition [38]. During the treatment, breast cancer patients face daily routine changes starting with the chemo-infusion, which can be done in an outpatient setting or, depending on the case, during hospitalization [4]. Food intake and physical activity can also be changed due to side effects such as fatigue, reduced immunity, and severe anemia [42], which lead to less energy flow and consequent slowdown in metabolism [36,38]. For breast cancer patients, results reinforce the need to assess body components, as well as to consider the body fat disposition [8] and the presence of sarcopenia [42].
Interestingly, our study found an increase in FFMI in women who were using the AC regime through the post-hoc Sidak test. On the other hand, Del Rio et al. [43] found a significant increase in fat-free mass after the CMF regimen (43.6 ± 1.3 vs. 45.2 ± 1.5 kg, p < 0.001), which could be explained by the fluid retentions during the treatment that are conducive to a body water increase, and not a higher free-fat mass [19,44].
A potential limitation of our study was the use of non-standardized methods to assess body composition. Another limitation would be the lack of specific variables not collected in some centers (e.g., WC and WHtR), which resulted in a smaller sample size in certain analyses. Our study has some strengths. Firstly, the patient was included as a random variable in the GMM and the maximum time to complete CT (6 months) has been standardized. Secondly, the originality of the study and the involvement of different Brazilian regions allows a broader knowledge of the sociodemographic, clinical, hormonal, and therapeutic characteristics of Brazilian patients with breast cancer. Finding that there is an effect of weight gain and body composition during chemotherapy treatment and the impact of the regimes used was also key, highlighting that management weight recommendations should be considered during this treatment phase.

Conclusions
Our study concluded that there was an increase in BMI and FMI after CT, and the impact on FMI and FFMI was dependent on the CT regimen. FMI significantly increased in women after ACT treatment, while FFMI increased after AC. These results are important to disseminate to the scientific community and health professionals in order to adjust diet and physical activity to counterbalance body composition changes caused by CT treatment in breast cancer patients. Funding: This research received funds to cover publication costs by Fundação Cearense de Amparo à Pesquisa (FUNCAP) (P20-0171-00041.01.00/20). The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Institutional Review Board Statement:
The study was conducted according to the guidelines of the Declaration of Helsinki and approved by the Ethics Committee from the Clinical Hospital of the Federal University of Goiás (n.3.858.331/2020). Written informed consent was provided by all participants after all risks, discomforts, and benefits involved in the study were reviewed in each region.
Informed Consent Statement: Informed consent was obtained from all subjects involved in the study.

Data Availability Statement:
The data presented in this study are available on request from the corresponding author.