Weight Loss During Neoadjuvant Therapy Is Associated With Poor Response Among the Patients With Gastrointestinal Cancer: A Propensity Score Matching Analysis

Abstract

Purpose

The aim of the current study was to identify the relationship between body composition changes during neoadjuvant therapy (NAT) and the treatment efficiency of NAT in gastrointestinal cancer (GC) patients.

Methods

From January 2015 to July 2020, 277 GC patients treated with NAT had included for retrospective analysis. The body mass index (BMI) and computed tomography (CT) imaging before and after NAT were recorded. The BMI change optimal cut-off value were calculated by ROC curve. Balancing essential characteristic variables using propensity score matching (PSM) method. Exploring the association between BMI changes and tumor response to NAT using logistic regression analysis. The survival outcome of matched patients between different BMI change groups was compared.

Results

A cutoff point of BMI change >2% during NAT was defined as BMI loss. Among the 277 patients, 110 (39.7%) patients showed BMI change with a loss after NAT. In total, 71 pairs of patients were selected for further analysis. The median follow-up time was 22 months (range 3 to 63 months). Univariate and multivariate logistic regression analyses in matched cohort showed that BMI change was a prognostic factor for tumor response after NAT in GC patients (odds ratio (OR), .471; 95% confidence interval (CI), .233-.953; P = .036). In addition, patients who experienced BMI loss after NAT showed worse overall survival than those who had BMI gain or stable.

Conclusion

BMI loss during NAT probably may has negative effects on NAT efficiency and survival for gastrointestinal cancer patients. It is necessary to monitor and maintain weight for patients during treatment.

Graphical Abstract

Keywords

gastrointestinal cancer body mass index neoadjuvant therapy skeletal muscle index survival

Introduction

Neoadjuvant treatment, including neoadjuvant chemo-or chemoradiotherapy, is increasingly used in malignancies of the gastrointestinal tract,^1–3 particularly gastric cancers and colorectal carcinomas. Neoadjuvant therapy (NAT) contributes substantially to improving outcomes and provides survival benefits for patients.^4,5

Gastrointestinal cancers that develop in the stomach and colorectum are common cancers and have a very poor prognosis.^6,7 Incredibly, 15% of patients are diagnosed with gastrointestinal carcinoma with distant organ metastasis, and liver metastases are the most common.^8–10 Radical resection offers the best opportunity for cure in gastrointestinal cancer (GC), but a great majority of patients with unresectable gastrointestinal cancer in their first diagnosis or surgery cannot obtain many benefits.

Multidisciplinary therapy has been well implemented in the treatment of locally advanced GC or metastases.^11–13 The response of primary malignant tumors or metastasis after neoadjuvant chemo- or radiochemotherapy is usually assessed by computed tomography (CT) or magnetic resonance imaging (MRI). In addition, tumor regression grade (TRG) can be used to estimate resection specimens, which enables comparisons of interstitial fibrosis and residual tumor tissue.^14,15 Evidence suggests that GC patients experiencing weight loss, malnutrition, and muscle wasting during neoadjuvant chemo- or chemoradiotherapy have poor survival prognoses.^16,17 Changes in body composition may also reflect the effect of antitumor therapy. Nevertheless, data on the efficacy of neoadjuvant therapy are lacking, which could be used to preliminarily evaluate the nutritional status of patients.

Several body components, as well as body mass index (BMI), skeletal muscle index (SMI) and muscle characteristics, have been observed to be correlated with hospitalization, surgery and survival in non-cancer patients.^18–20 The evaluation of body composition using cross-sectional CT imaging has become popular due to its wide availability and high precision. BMI is a prognostic indicator for assessing patients’ risk of malnutrition.²¹ Meanwhile, the evaluation of muscle data is an emerging area for improving preoperative risk factors.²²

In our study, body weight, skeletal muscle area (SMA) and changes were determined by BMI, SMI, BMI change and SMI change. The ultimate goal of our study was to expound on the correlation between body composition and the effectiveness of neoadjuvant therapy in patients with GC and to investigate whether BMI change and SMI change could predict response and/or overall survival (OS) after neoadjuvant treatment.

Materials and Methods

Patient Selection

This retrospective analysis was conducted in GC patients who received NAT from January 2015 to July 2020 at our center. The course of NAT was determined based on the two different physicians’ choices in each patient. After NAT was accomplished, most patients underwent radical surgery. The included patients had histologically confirmed gastrointestinal cancer with complete BMI and CT imaging data before and after NAT. Mortality data were collected at follow-up. Adverse events were classified according to the Common Terminology Criteria for Adverse Events.²³ Necessary medical care was conducted on patients for adverse events of 3 or 4 grades, including supportive treatments, blood transfusion, and entire rest. No cachexia in any patient. All patient details have de-identified in this study.

The essential clinicopathological characteristics of the patients included age, sex, weight, BMI, SMA, SMI, hemoglobin, albumin, tumor markers, NAT treatment, pathological tumor characteristics and radical surgery type. Patients received systemic therapy as a multidisciplinary choice. The survival status of GC patients was reviewed by medical records and phone calls. The reporting of this study conforms to STROBE guidelines.²⁴

Ethics Statement

All patients signed written informed consent and allowed their data to be disclosed. This study protocol was reviewed and approved by the Ethics and Human Subject Committee of our center (LW2022181). All materials and methods were performed according to relevant guidelines and regulations.

Body Composition Measurements

Body weight was evaluated with BMI as a measure of obesity, dividing the square of height (m) into body weight (kg), which was sorted into categories according to the World Health Organization: underweight, which was judged as BMI <24 kg/m²; overweight, which was judged as 24.0 to 29.9 kg/m²; and obesity, which was judged as ≥ 30 kg/m². BMI change was defined as fluctuation between Day 1 of the first cycle and the last cycle of NAT or the day before the operation of GC patients.

SMA was calculated before and after NAT in GC patients using high-resolution CT scans and analyzed using ImageJ (ImageJ software, https://imagej.nih.gov/ij) for precise measurements. The SMA (in cm²), calculated within a Hounsfield unit range of −29 to 150 at the level of the third lumbar vertebra of all the muscle tissues (including the rectus abdominis, intra-abdominal oblique muscle, external oblique muscle, transverse abdominis muscle, paraspinal muscle and psoas muscle), was summed for each image (Figure S1). Then, SMI is the muscle mass index obtained by dividing SMA (cm²) by the square of the patient’s height (m²). The change in SMI was determined between the initial data and after the second cycle of NAT. Sarcopenia is defined as an SMI lower than 34.9 cm²/m² in women or lower than 40.8 cm²/m² in men, as described by Zhuang et al.²⁵ The radiological evaluation was done by two trained professionals who were blinded to the clinical diagnosis at the time of the calculation. If the results were not agreed upon, a third physician evaluated the patients.

Treatment Efficacy Evaluation

Before and after NAT, patients received enhanced CT scans. The efficiency of NAT was judged according to Resist1.1²⁶ evaluation criteria: complete response (CR): all target lesions have disappeared, any pathological lymph nodes must have a reduction to <10 mm in the shortest axis; partial response (PR): at least 30% decline in the total diameters of target lesions, which is taken as reference for the baseline diameters; stable disease (SD): neither sufficient shrinkage to qualify for PR nor sufficient increase to qualify for progressive disease (PD), defined as a stable situation; PD: at least 20% augmentation in the sum total of diameters of target lesions, which is taken as a reference for the smallest sum in the study. In addition to the proportionable boost of 20%, the sum must have an absolute boost of at least 5 mm. SD and PD are considered nonresponders undergoing preoperative therapy.

NCCN guidelines recommend using a TRG system to evaluate tumor response to treatment.²⁷ Therefore, for patients undergoing radical resection, we also used the TRG system to assess the efficacy of NAT. TRG0: no residual tumor cells; TRG1: single cells or small groups of cells; TRG2: residual cancer with desmoplastic response; and TRG3: minimal evidence of tumor response. TRG0 is considered a pathologic complete response (pCR), and TRG1, TRG2, and TRG3 are considered none of the pathologic complete responses (npCRs). Two comparative histopathological analyses were performed for postoperative tissue sections after NAT.

Statistical Analyses

All optimal cut-off values were calculated by ROC curves. OS was defined as the time interval from the date of the first cycle of NAT to the date of death from any cause. The survival outcomes were assessed by log-rank test and Kaplan-Meier analysis. Multivariate Cox analysis was performed for potential independent prognostic factors with P < .05 and selected in the univariable analysis. (Kaplan–Meier curves were used to compare survival in the different groups with the log-rank test (by the “survival” package in R). Univariate and multivariate Cox proportional hazards regression analyses were used to calculate hazard ratio (HR) and 95% CI for clinicopathological variables and survival.)

Propensity score matching (PSM) analysis was performed to match patients using the nearest-neighbor method with a caliper of .05. Matching was based upon age, gender, primary cancer site, T stage, N stage, liver metastasis, clinical stage, NAT treatment, SMI change, tumor markers, hemoglobin, albumin, the efficiency of NAT, pathologic complete response after surgery. According to the cutoff value of median wait time, patients were classified as BMI-gaining/stable group and BMI-loss group as outlined above. Univariate and multivariable logistic regression analyses were used to calculate the HR and 95% CI of different clinicopathological variables for the efficiency of NAT. Variables with a P value of <.05 were included in the subsequent multivariate (logistic regression or Cox proportional hazards regression) analysis.

Univariate and multivariate Cox proportional hazards modeling was done on the propensity-matched sample. The Chi-square test or Fisher’s exact test was adopted to compare the differences of patients’ baseline characteristics.

All reported P values were 2-sided, and P < .05 was considered to be statistically significant. All statistical analyses were computed using IBM SPSS 25.0 and R software (version 3.6.3).

Results

Essential Patient Characteristics and Associations With BMI Change Categories

Between January 2015 and July 2020, 575 patients with stomach or colorectal carcinoma of the gastrointestinal was diagnosed at our center, then their course of NAT was accomplished. Of these, 277 (48.17%) met our cohort definition and were included in the analysis. A BMI gain or loss >2% following NAT was interpreted as significant, while changes in BMI ranging from −2% to 2% were interpreted as stable based on ROC analysis results. Patients were included either into the BMI-loss group (BMI loss >2%) or into the BMI-gaining/stable group (BMI changes in ±2% or BMI gaining >2%) according to the BMI changes to compare during NAT. The clinicopathological characteristics according to BMI-change classification are listed in Table 1. Primary cancer site, post neoadjuvant therapy (post-) weight, pretreatment (pre-) BMI, post-BMI, pre-CA125 level, and pre-Hb level differed between BMI-gaining/loss patients. Other clinicopathological characteristics were similar between the two groups.

Table 1.

Essential Patient Characteristics Stratified by BMI Change (n = 277).

Characteristics	NO. (%)	BMI Change		P-value
Characteristics	x ± s	BMI-gain/stable (n = 167)	BMI-loss (n = 110)	P-value
Age (y)	55 ± 12	54 ± 12	56 ± 12	.070
Gender				.175
Male	207 (74.7)	120 (58.0)	87 (42.0)
Female	70 (25.3)	47 (67.1)	23 (32.9)
Primary cancer site				<.01
Stomach	82 (29.6)	41 (50.0)	41 (50.0)
Colon	40 (14.4)	15 (37.5)	25 (62.5)
Rectum	155 (56.0)	111 (71.6)	44 (28.40)
T Stage				.527
T2	4 (1.4)	2 (1.2)	2 (1.8)
T3	91 (32.9)	59 (35.3)	32 (29.1)
T4	182 (65.7)	106 (63.5)	76 (69.1)
N stage				.463
N0	40 (14.4)	23 (13.8)	17 (15.5)
N1	152 (54.9)	92 (68.9)	60 (54.5)
N2	81 (29.2)	52 (31.1)	29 (26.4)
N3	4 (1.4)	0 (0)	4 (3.6)
Liver metastasis				.349
Yes	29 (10.5)	13 (7.8)	16 (14.5)
No	248 (89.5)	154 (92.2)	94 (85.5)
TNM stage				.745
I	2 (.7)	1 (.6)	1 (.9)
II	35 (12.6)	20 (12.0)	15 (13.6)
III	212 (76.5)	134 (80.2)	78 (70.9)
IV	28 (10.1)	12 (7.2)	16 (14.6)
Neoadjuvant therapy				.096
Chemotherapy alone	226 (81.6)	131 (60.0)	95 (40.0)
Chemotherapy + radiotherapy	51 (18.4)	36 (70.6)	15 (29.4)
Radical surgery				.450
Yes	247 (89.2)	147 (59.5)	100 (40.5)
No	30 (10.8)	20 (66.7)	10 (33.3)
Pre-weight	59.0 ± 9.6	58.8 ± 9.3	59.3 ± 10.1	.718
Post-weight	58.8 ± 9.5	60.7 ± 8.9	55.9 ± 9.6	<.01
Pre-BMI	22.18 ± 3.16	21.66 ± 2.99	22.97 ± 3.25	<.01
Post-BMI	22.16 ± 3.35	23.03 ± 3.02	20.84 ± 3.42	<.01
BMI-change	.45 ± 12.08	6.84 ± 9.93%	−9.26 ± −7.84%
Pre-SMA	104.50 ± 27.43	102.38 ± 27.16	107.63 ± 27.65	.860
Post-SMA	103.13 ± 26.25	101.94 ± 25.31	104.93 ± 27.65	.250
Pre-SMI	38.66 ± 10.65	37.76 ± 11.25	40.04 ± 9.57	.221
Post-SMI	38.73 ± 9.15	38.46 ± 8.56	39.14 ± 10.00	.162
Pre-CEA (ng/mL)	18.85 ± 44.83	16.30 ± 37.28	22.73 ± 54.29	.279
Pre-CA125 (ng/mL)	14.6 ± 23.52	11.67 ± 8.94	19.11 ± 35.29	.032
Pre-hb (g/L)	121 ± 32	124 ± 35	116 ± 27	.041
Pre-ALB (g/L)	36.7 ± 3.9	36.8 ± 4.0	36.6 ± 3.9	.721

Note: Values are presented as number (%).

Abbreviations: BMI, body mass index; SMA, skeletal muscle area; SMI, skeletal muscle index; T stage, tumor stage; N stage, node stage; Hb, hemoglobin; ALB, albumin; Pre_, pretreatment; Post-, post neoadjuvant chemotherapy.

In total, 277 patients who received NAT were included in the final analysis, including 82 (29.6%) patients with stomach cancer, 155 (56.0%) with rectal cancer, and 40 (14.4%) with colon cancer. In most patients (81.6%), NAT was based on chemotherapy alone, with the remaining patients receiving chemoradiotherapy. Most of the GC patients (89.2%) received radical surgery. The mean BMI pretreatment for all patients was 22.18 ± 3.16 kg/m². The mean BMI after NAT was 22.16 ± 3.35 kg/m², and the mean BMI change was .45 ± 12.08%. Meanwhile, the average SMI of all patients before NAT was 38.66 ± 10.65 cm²/m², and the average SMI after NAT was 38.73 ± 9.15 cm²/m².

PSM analysis was done with the same dataset in GC patients for whom with BMI-gaining/stable and BMI-loss. Of these 110 patients, who with BMI loss during neoadjuvant treatment, 71 (64.5%) were matched successfully. Matched variables from essential patient characteristics were adequately balanced based on predetermined criteria. (Table 2).

Table 2.

Baseline Characteristic of Patients Stratified by BMI Change Before and After Matching.

Characteristics	Pre-matching			Post-matching
Characteristics	BMI-Gain/Stable(n = 167)	BMI-Loss(n = 110)	P-value	BMI-Gain/Stable (n = 71)	BMI-Loss (n = 71	P-value
Age (years)			.074			1.000
≤60	115 (68.9)	64 (58.2)		28 (39.4)	27 (38.0)
>60	52 (31.1)	46 (41.8)		43 (60.6)	44 (62.0)
Gender			.204			.677
Male	120 (71.9)	87 (79.1)		55 (77.5)	58 (81.7)
Female	47 (28.1)	23 (20.9)		16 (22.5)	13 (18.3)
Primary cancer site			<.001			.194
Stomach	41 (24.6)	42 (38.2)		27 (38.0)	24 (33.8)
Colon	15 (9.0)	24 (21.8)		5 (7.0)	12 (16.9)
Rectum	111 (66.5)	44 (40.0)		39 (55.0)	35 (49.3)
T Stage			.527			.707
T2	2 (1.2)	2 (1.8)		0 (0)	0 (0)
T3	59 (35.3)	32 (29.1)		18 (25.4)	21 (29.6)
T4	106 (63.5)	76 (69.1)		53 (74.6)	50 (70.4)
N stage			.081			.257
N0	23 (13.8)	17 (15.5)		7 (9.9)	10 (14.1)
N1	92 (55.1)	60 (54.5)		37 (52.1)	40 (56.3)
N2	52 (31.1)	29 (26.4)		27 (38.0)	19 (26.8)
N3	0 (0)	4 (3.6)		0 (0)	2 (2.8)
Liver metastasis			.107			.743
No	154 (92.2)	94 (85.5)		65 (91.8)	67 (94.4)
Yes	13 (7.8)	16 (14.5)		6 (8.5)	4 (5.6)
Clinical stage			.209			.552
I	1 (.6)	1 (.9)		0 (0)	0 (0)
II	20 (12.0)	15 (13.6)		6 (8.5)	10 (14.1)
III	134 (80.2)	78 (70.9)		60 (84.5)	57 (80.3)
IV	12 (7.2)	16 (14.5)		5 (7)	4 (5.6)
Neoadjuvant therapy			.217			.685
Chemotherapy alone	131 (78.4)	93 (84.5)		54 (76.1)	57 (80.3)
Chemotherapy + radiotherapy	36 (21.6)	17 (15.5)		17 (23.9)	14 (19.7)
Radical surgery			1.000			1.000
No	10 (6.0)	6 (5.5)		4 (5.6)	3 (4.2)
Yes	157 (94.0)	104 (94.5)		67 (94.4)	68 (95.8)
SMI change			.390			.737
Loss	83	61		35	38
Gain	84	49		36	33
CEA (ng/mL)			.176			.867
≤5	72 (43.1)	57 (51.8)		35 (49.3)	33 (46.5)
>5	95 (56.9)	53 (48.2)		36 (50.7)	38 (53.5)
CA125 (ng/mL)			.010			1.000
≤35	5 (3.0)	12 (10.9)		2 (2.8)	2 (2.8)
>35	162 (97.0)	98 (89.1)		69 (97.2)	69 (97.2)
Hb (g/L)			.235			.828
≤90	22 (13.2)	21 (19.1)		12 (16.9)	14 (19.7)
>90	145 (86.8)	89 (80.9)		59 (83.1)	57 (80.3)
ALB (g/L)			.892			1.000
≤35	47 (28.1)	32 (29.1)		19 (26.8)	19 (26.8)
>35	120 (71.9)	78 (70.9)		52 (73.2)	52 (73.2)
Efficiency of NAT			.049			.053
CR/PR	62 (37.1)	28 (25.5)		31 (43.7)	19 (26.8)
SD/PD	105 (62.9)	82 (75.5)		40 (56.3)	52 (73.2)
Pathologic complete response after surgery			.230			.452
NA	20 (12.0)	21 (19.1)		7 (9.9)	12 (16.9)
pCR	20 (12.0)	10 (9.1)		12 (17.0)	10 (14.1)
Non-pCR	127 (76.0)	79 (71.8)		52 (73.2)	49 (69.1)

Note: Values are presented as number (%).

Abbreviations: T stage, tumor stage; N stage, node stage; Hb, hemoglobin; ALB, albumin. pCR, pathologic complete response.

NA: Thses patients did not conduct surgery.

The Effect of BMI-Change on NAT Response

By comparison, CR/PR is considered to have good NAT efficacy, and SD/PD is deemed to have poor NAT efficacy according to RECIST1.1 criteria (23). A total of 32.4% (90/277) of patients acquired CR/PR following NAT. Patients with CR/PR comprised 25.3% (75/277) of the underweight group; those in the overweight/obesity group were similar (24.1%, 27/112) but different from the BMI-gaining/stable (23.3%) and BMI-loss (25.1%) groups. Within the propensity-matched cohort, the multivariable logistic regression analyses identified BMI change as a prognostic factor for CR/PR in GC patients who underwent NAT (Table 3). However, SMI change was not predictive of efficacy after NAT in the matched cohort.

Table 3.

Univariate and Multivariate Logistic Regression Analyses of Prognostic Factors for Tumor Response of CR/PR to NAT in the Matched Cohort (n = 142).

Variable	Univariate Analysis		Mulivariate Analysis
Variable	OR (95% CI)	P-value	OR (95% CI)	P-value
Age (>60 vs ≤ 60)	1.086 (.536-2.198)	.819
Gender (male vs Female)	1.553 (.632-3.817)	.337
Primary cancer site (colon vs Stomach)	1.778 (.582-5.428)	.312
Primary cancer site (rectum vs Stomach)	1.020 (.479-2.172)	.958
T Stage (T4 vs T2/3)	1.841 (.810-4.184)	.145
N stage (N2-3 vs N0-1)	1.014 (.490-2.098)	.971
Liver metastasis (yes vs No)	3.000 (.805-11.184)	.102
Neoadjuvant therapy (Chemotherapy + radiotherapy vs Chemotherapy alone	.845 (.362-1.971)	.697
SMI change (loss vs Gain)	.917 (.460-1.826)	.804
CEA (>5 vs ≤ 5)	.685 (.343-1.368)	.283
CA125 (>35 vs ≤ 35)	1.652 (.167-16.308)	.668
Hb (>90 vs ≤ 90)	.842 (.350-2.026)	.701
ALB (>35 vs ≤ 35)	1.247 (.565-2.754)	.584
BMI-change (loss vs Gain/stable)	.471 (.233-.953)	.036	.471 (.233-.953)	.036

Overall,10.8% (30/277) of patients received pCR after surgery. The pCR ratio was similar to that of the BMI-gaining/stable (13.6%, 20/147) group. After matching, none of the listed clinicopathological variables were correlated with pCR (Table 4). The primary cancer site of GC patients did not seem to correlate with the response rate of NAT.

Table 4.

Univariate and Multivariate Logistic Regression Analyses of Prognostic Factors for Tumor Response of pCR to Surgery in the Matched Cohort (n = 123).

Variable	Univariate Analysis		Mulivariate Analysis
Variable	HR (95% CI)	P-value	HR (95% CI)	P-value
Age (>60 vs ≤ 60)	.988 (.379-2.577)	.981
Gender (male vs Female)	1.764 (.478-6.511)	.395
Primary cancer site (colon vs Stomach)	NA	.998
Primary cancer site (rectum vs Stomach)	1.471 (.544-3.977)	.447
T Stage (T4 vs T2/3)	.576 (.216-1.533)	.269
N stage (N2-3 vs N0-1)	.807 (.302-2.160)	.670
Liver metastasis (yes vs No)	NA	.999
Neoadjuvant therapy (Chemotherapy + radiotherapy vs Chemotherapy alone)	4.316 (1.631-11.423)	.003
SMI change (loss vs Gain)	.817 (.324-2.061)	.669
CEA (>5 vs ≤ 5)	1.020 (.406-2.565)	.966
CA125 (>35 vs ≤ 35)	.210 (.013-3.493)	.688
Hb (>90 vs ≤ 90)	1.181 (.361-3.864)	.783
ALB (>35 vs ≤ 35)	.597 (.216-1.645)	.318
BMI-change (loss vs Gain/stable)	.884 (.351-2.231)	.795

Note: Values are presented as number (%).

Abbreviations: T stage, tumor stage; N stage, node stage; Hb, hemoglobin; ALB, albumin.

The Effect of BMI-Change on Overall Survival

The median follow-up for the entire study population was 22 months (3 to 63 months), and that for the survivors was 19 months (3 to 63 months). During follow-up, 47 deaths had been listed until after the last follow-up, 45 patients died of gastrointestinal cancers, and two died of other reasons. Patients with BMI gain/stability post NAT in the matched cohort had longer-term survival than patients with BMI loss by Kaplan–Meier analysis (P = .043, Figure 1). The SMI change did not correlate with the NAT response. However, SMI change was closely reached statistical significance, which means it may be a potential prognosis factor in a larger population. (P = .144, Table 5). Univariate analysis identified primary cancer site, Hb, and BMI change type as potential risk factors for OS in patients who received NAT. BMI change type was independent prognostic factors for OS (HR, 2.365; 95% CI, 1.067-5.240; P = .034).

Figure 1.

Kaplan-Meier overall survival analysis of GC patients based on BMI-change after propensity score matching.

Table 5.

Univariate and Multivariate Cox Analysis of Prognostic Factors for Overall Survival in the Matched cohort (n = 142).

Variable	Univariate Analysis		Mulivariate Analysis
Variable	HR (95% CI)	P-value	HR (95% CI)	P-value
Age (>60 vs ≤ 60)	1.013 (.454-2.257)	.976
Gender (male vs Female)	3.665 (.868-15.480)	.077
Primary cancer site (colon vs Stomach)	.543 (.160-1.845)	.103	.451 (.132-1.547)	.206
Primary cancer site (rectum vs Stomach)	.186 (.069-.499)	.001	.198 (.072-.542)	.002
T Stage (T4 vs T2/3)	1.043 (.441-2.469)	.923
N stage (N2-3 vs N0-1)	1.010 (.462-2.210)	.980
Liver metastasis (yes vs No)	2.158 (.639-7.287)	.215
Neoadjuvant therapy (Chemotherapy + radiotherapy vs Chemotherapy alone	.376 (.113-1.F249)	.110
Radical surgery (yes vs No)	.882 (.119-6.534)	.902
SMI change (loss vs Gain)	1.821 (.815-4.067)	.144
CEA (>5 vs ≤ 5)	1.471 (.671-3.223)	.335
CA125 (>35 vs ≤ 35)	.450 (.105-1.924)	.281
Hb (>90 vs ≤ 90)	.429 (.187-.985)	.046	.605 (.257-1.424)	.250
ALB (>35 vs ≤ 35)	.592 (.266-1.320)	.200
BMI-change (loss vs Gain/stable)	2.191 (1.004-4.781)	.049	2.365 (1.067-5.240)	.034

Note: Values are presented as number (%).

Abbreviations: T stage, tumor stage; N stage, node stage; Hb, hemoglobin; ALB, albumin.

Of these 142 patients, we performed a subgroup analysis of whom diagnosed with stomach cancer and colorectal cancer. The BMI loss had shown in matched colorectal patients with negative survival. (Figure S2) The findings of this study suggest that it is no difference in OS between patients with stomach cancer and colon cancer. But rectum cancer has a better prognosis than gastric cancer (Figure 2). In addition, Hb status during NAT was associated with OS. GC patients with non-anemia showed long-term survival during NAT (Figure 3).

Figure 2.

Kaplan-Meier overall survival analysis based on pretreatment Hb status within 71 pairs patients after matching.

Figure 3.

Kaplan-Meier overall survival analysis based on Primary cancer site within 71 pairs patients after matching.

The optimum cutoff points of some clinicopathological variables were determined by ROC curve analysis. Because our primary survival endpoint was OS in this study, the optimum cutoff points of age, pre-BMI, post-BMI, BMI change rate, pre-CEA, pre-CA125 and pre-Hb, pre-ALB for OS were 60 y, 20.55 kg/m2, 22.5 kg/m2, -2%, 7.93 U/ml, 10.3 U/ml, 134 g/L and 36.7 g/L, respectively.

Discussion

In this retrospective observational study of patients with GC who had neoadjuvant chemo- or chemoradiotherapy followed by radical resection for gastric or colorectal cancer, BMI loss after NAT in our cohort negatively affected the curative effect. However, GC patients’ appetite and weight changes were affected by different gastrointestinal tumor sites. So, we performed a subgroup analysis of whom diagnosed with stomach cancer and colorectal cancer. The results showed that there was no statistical difference between BMI change and NAT response rate based on different cancer types in the matched cohort. However, because of the small sample size after matching, and only 8 deaths included, we think that change in BMI did not seem to correlate with OS in the subgroup of patients with colorectal cancer.

The patients with severely reduced BMI had an increase compared to those with mild-to-moderate BMI. BMI loss during NAT may be an indispensable predictor of curative effects. Low SMI during NAT had a potential poor impact on OS, but it did not predict a weak response to NAT. Furthermore, we found that BMI loss (≥2%), but not SMI loss during neoadjuvant chemo- or chemoradiotherapy of GC patients, predicted significantly poor response and OS.

Weight loss is a common problem among GC cancer patients. Massive weight loss among cancer patients may be attributed to cachexia or sarcopenia. Most prior studies on the body composition of cancer patients were performed in Western countries, and their studies were incompatible. However, BMI loss during NAT can negatively affect survival, as previously reported.²⁸ In our study, a BMI reduction of 2% or more, corresponding to a weak response, was observed in 39.7% of patients during NAT and in 52.0% when patient-reported SMI loss was included. We identified patients with BMI-gaining/stable and the other with BMI-loss undergone NAT, and we adjusted for differences in baseline characteristics using propensity score matching. In the matched cohort, this study still revealed that GC patients during NAT benefited from BMI-gaining/stability with a significant response rate as well as overall survival.

In most recent studies, patients diagnosed with GC could benefit greatly from neoadjuvant therapy, and their local tumor control rate and overall survival were significantly improved.^29–31 However, this study focused on patients with resectable disease, with a selection of operative timing after neoadjuvant chemotherapy remaining controversial. For these patients, we hope to provide a more accessible and earlier predictor for the response to neoadjuvant treatment and survival than imaging. Precise medication can currently reduce the complications and occurrence of tumor-associated cachexia for cancer patients during neoadjuvant treatment.^32,33 BMI changes after NAT can be used as a basis, combined with imaging assessments and biochemical indicators, to evaluate the efficacy of neoadjuvant therapy in GC patients before surgery. Compared with previously published reports, a decline in BMI was positively associated with preoperative treatment efficacy and survival in our cohort. However, our data showed the relationship between changes in BMI and the efficacy or prognosis of gastric and colorectal cancer patients who underwent NAT in general compared to previously published studies. We emphasize that patients with locally advanced gastric and colorectal cancer, as well as patients with liver metastases from colorectal cancer who underwent surgery, were included in our study.

The decrease in weight and muscle mass during preoperative treatment may be related to nutritional consumption of digestive system neoplasms, and tumor-induced tests also lead to weight loss. The available data on weight loss related to NAT response, as well as BMI, have been controversial. Some studies have suggested that BMI gain is good for OS, which supports our data,²⁸ while others have shown contradictory results.³⁴ Although some studies have suggested that low BMI is a risk factor for cardiopulmonary complications, others have come to opposite conclusions and even reported that obesity or a high BMI is a risk factor for outcomes.³⁵ Daniel’s team found that BMI is a key factor affecting the prognosis of rectal cancer patients, wherein BMI >25 kg/m2 is significantly correlated with longer OS and Cancer specific survival.³⁶

Pathologic complete response is defined as the highest efficacy of neoadjuvant treatment at the time of surgery, and pCR represents an essential evaluation for NAT outcome.³⁴ In this study, 10.8% (30/277) of patients acquired pCR during preoperative treatment. All patients who were evaluated as the BMI-stable group were observed, and more than half (54%) of GC patients had a gain in BMI. The pCR percentage was similar to that of the BMI-loss group (11.2%, 10/89) and the BMI-gain/stable (13.6%, 20/147) group (P = .653), which is consistent with most reports. However, BMI change was a potential independent prognostic factor for poor OS, while BMI loss could project poor response to NAT and poor OS. This is the first study to investigate the correlation among BMI change during NAT, pCR percentage and survival of Chinese patients with gastrointestinal cancer.

There is growing evidence recently that SMA can predict response in patients undergoing preoperative treatment and that these assessments are more accurately than overall changes in weight or BMI.^11,37 In our data, SMI change did not correlate with the NAT response, but it may be a potential prognosis factor in a larger population because it was closely reached statistical significance in the matched cohort. In particular, loss of muscle mass is a predicted indicator of adverse effects, lengthy postoperative recovery, low adjuvant chemotherapy tolerance, and poorer long-term survival. It is well known that the progressive depletion of SMA during NAT has been associated with poor postoperative outcomes in digestive system cancers, such as esophageal and gastric cancers.^11,38 Skeletal muscle mass loss and BMI gain may increase the drug toxicity of neoadjuvant chemotherapy, which is more likely to cause NAT complications and affect the subsequent treatment of GC patients.

Despite the large sample size before matching, this retrospective study has limitations. Our single-center studies with limited samples may have resulted in biases in the analysis, and multicenter studies with larger samples are needed to investigate this further. In addition, the modalities of neoadjuvant therapy differed in this study, including neoadjuvant chemotherapy alone, neoadjuvant radiotherapy alone, and neoadjuvant concurrent chemoradiotherapy. No subgroup analysis was performed. Future studies should focus on the etiology of body composition changes in patients with cancer.

Conclusions

This study revealed that poor response and OS after neoadjuvant therapy may associated with significant weight loss in GC patients. We suggest that BMI-change can be a more accessible and earlier predictor for the response to neoadjuvant treatment and survival in GC patients than imaging due to its clinical practicality. This could prompt clinicians to take timely measures to interfere with nutrition during NAT.

Supplemental Material

Supplemental Material - Weight Loss During Neoadjuvant Therapy Is Associated With Poor Response Among the Patients With Gastrointestinal Cancer: A Propensity Score Matching Analysis

Supplemental Material for Weight Loss During Neoadjuvant Therapy Is Associated With Poor Response Among the Patients With Gastrointestinal Cancer: A Propensity Score Matching Analysis by Zhaoting Bu, Yuting Jiang, Shanshan Luo, Xinxin He, Haiquan Qin, and Weizhong Tang in Cancer Control

Footnotes

Acknowledgments

We thank Yongsheng Meng, Jianhong Chen, Chao Tian, and Xiaoliang Huang for their efforts in establishing the database. We would like to thank Hui Long for English language grammar review.

Author Contributions

Conceptualization, SL. Writing-original draft preparation, ZB, YJ and SL. Writing-review and editing, ZB, XH, HQ. Supervision, YJ, WT. All authors contributed to the article and approved the submitted version.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Guangxi Science and Technology Project (AD19245197), the Guangxi Scientific Research and Technology Development Project (Guike AB18126033), Research Basic Ability Improvement Project for Guangxi Young College Teachers (2021KY0087), the Innovation Project of Guangxi Graduate Education (YCSW2022217), and the Natural Science Foundation of Guangxi Province (2018GXNSFAA294013).

Ethical Approval

The studies involving human participants were reviewed and approved by the Ethics Committee of Guangxi Medical University Cancer Hospital (LW2022181). Since this study was retrospective, the ethics committee waived the requirement of written informed consent for participation. Written informed consent was not obtained from the individual(s) for the publication of any potentially identifiable images or data included in this article.

Data Availability

All original data included in the manuscript are available upon reasonable request.

ORCID iD

Zhaoting Bu

Supplemental Material

Supplemental material for this article is available online.

Appendix

References

Shan

Ying

, et al. Assessment of laparoscopic distal gastrectomy after neoadjuvant chemotherapy for locally advanced gastric cancer: A randomized clinical trial. JAMA Surg. 2019;154:1093-1101.

Van der Werf

Dikken

van der Willik

, et al. Time interval between neoadjuvant chemoradiotherapy and surgery for oesophageal or junctional cancer: A nationwide study. Eur J Cancer. 2018;91:76-85.

Costas-Chavarri

Nandakumar

Temin

, et al. Treatment of patients with early-stage colorectal cancer: ASCO resource-stratified guideline. J Glob Oncol. 2019;5:1-19.

Busweiler

LAD

Jeremiasen

Wijnhoven

BPL

, et al. International benchmarking in oesophageal and gastric cancer surgery. BJS Open. 2019;3:62-73.

Boughanem

Martin-Nuñez

Torres

, et al. Impact of tumor LINE-1 methylation level and neoadjuvant treatment and its association with colorectal cancer survival. J Personalized Med. 2020;10:219.

Bray

Ferlay

Soerjomataram

Siegel

Torre

Jemal

. Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin. 2018; Nov;68(6):394-424.

Dienstmann

Vermeulen

Guinney

Kopetz

Tejpar

Tabernero

. Consensus molecular subtypes and the evolution of precision medicine in colorectal cancer. Nat Rev Cancer. 2017; Feb;17(2):268-292.

Kow

AWC

. Hepatic metastasis from colorectal cancer. J Gastrointest Oncol. 2019; Dec;10(6):1274-1298.

Yamada

Ozawa

Onozato

Suzuki

Fujita

Ojima

. Optimal timing for the resection of pulmonary metastases in patients with colorectal cancer. Medicine (Baltim). 2020; Feb;99(9):e19144.

10.

Canda

Arslan

Terzi

, et al. Treatment of intraoperatively detected peritoneal carcinomatosis of colorectal origin with cytoreductive surgery and intraperitoneal chemotherapy. World J Surg Oncol. 2018; Mar 27;16(1):70.

11.

Aoyama

. Perioperative body composition changes in the multimodal treatment of gastrointestinal cancer. Surg Today. 2020; Mar;50(3):217-222.

12.

Neumann

Berlet

Friess

. Surgical oncology in the age of multimodality therapy for cancer of the upper and lower gastrointestinal tract. Expert Rev Anticancer Ther. 2021; 21:511-522.

13.

Joharatnam-Hogan

Wilson

Shiu

, et al. Multimodal treatment in metastatic colorectal cancer (mCRC) improves outcomes-the university college London hospital (UCLH) experience. Cancers. 2020; Nov 27;12(12):3545.

14.

Langer

Becker

. Tumor regression grading of gastrointestinal cancers after neoadjuvant therapy. Virchows Arch. 2018; Feb;472(2):175-186.

15.

Tong

Liu

Zhang

. Connection and distinction of tumor regression grading systems of gastrointestinal cancer. Pathol Res Pract. 2020;216:153073.

16.

Lobo

Gianotti

Adiamah

, et al. Perioperative nutrition: Recommendations from the ESPEN expert group. Clin Nutr. 2020; Nov;39(11):3211-3227.

17.

Deftereos

Kiss

Isenring

Carter

Yeung

. A systematic review of the effect of preoperative nutrition support on nutritional status and treatment outcomes in upper gastrointestinal cancer resection. Eur J Surg Oncol. 2020; Aug;46(8):1423-1434.

18.

Cioffi

Imperatore

Di Vincenzo

, et al. Association between health-related quality of life and nutritional status in adult patients with Crohn’s disease. Nutrients. 2020; Mar 11;12(3):746.

19.

Singh

Facciorusso

Singh

, et al. Obesity and response to anti-tumor necrosis factor-α agents in patients with select immune-mediated inflammatory diseases: A systematic review and meta-analysis. PLoS One. 2018; May 17;13(5):e0195123.

20.

Szilagyi

. Relationship(s) between obesity and inflammatory bowel diseases: Possible intertwined pathogenic mechanisms. Clin J Gastroenterol. 2020;13:139-152.

21.

Naumann

Eberlein

Farnia

Hackert

Debus

Combs

. Continued weight loss and sarcopenia predict poor outcomes in locally advanced pancreatic cancer treated with chemoradiation. Cancers. 2019; May 23;11(5):709.

22.

Yoon

Choi

Song

, et al. Impact of preoperative body compositions on survival following resection of biliary tract cancer. J Cachexia Sarcopenia Muscle. 2019; Aug;10(4):794-802.

23.

Hozumi

Hasegawa

Miyashita

, et al. Efficacy of corticosteroid and intravenous cyclophosphamide in acute exacerbation of idiopathic pulmonary fibrosis: A propensity score-matched analysis. Respirology. 2019; Aug;24(8):792-798.

24.

Von Elm

Altman

Egger

, et al. The strengthening the reporting of observational studies in epidemiology (STROBE) statement: Guidelines for reporting observational studies. Ann Intern Med. 2007;147:573-577.

25.

Zhuang

Huang

Pang

, et al. Sarcopenia is an independent predictor of severe postoperative complications and long-term survival after radical gastrectomy for gastric cancer: Analysis from a large-scale cohort. Medicine (Baltim). 2016; Mar;95(13): e3164.

26.

Benson

Venook

Al-Hawary

, et al. NCCN guidelines insights: Rectal cancer, version 6.2020. J Natl Compr Cancer Netw. 2020; Jul;18(7):806-815.

27.

Park

Kim

Joo

Lee

Han

. Important CT and histopathological findings for recurrence and overall survival in patients with pancreatic ductal adenocarcinoma who underwent surgery after neoadjuvant Folfirinox. Eur Radiol. 2021; Jun;31(6):3616-3626.

28.

Mantzorou

Koutelidakis

Theocharis

Giaginis

. Clinical value of nutritional status in cancer: What is its impact and how it affects disease progression and prognosis? Nutr Cancer. 2017;69(8):1151-1176.

29.

Zhong

Chen

Parisi

, et al. Modified ypTNM staging classification for gastric cancer after neoadjuvant therapy: A multi-institutional study. Oncol. 2021; Jan;26(1):e99-e110.

30.

Ravetch

Langdon-Embry

, et al. eds. The newly proposed clinical and post-neoadjuvant treatment staging classifications for gastric adenocarcinoma for the American joint committee on cancer (AJCC) staging. Gastric Cancer. 2018; Jan;21(1):1-9.

31.

Huang

, et al. Outcomes of neoadjuvant chemoradiotherapy followed by radical resection for T4 colorectal cancer. World J Gastrointest Oncol. 2020; Dec 15;12(12):1428-1442.

32.

Chan

Chin

Low

. The roles of common variation and somatic mutation in cancer pharmacogenomics. Oncol Ther. 2019; Jun;7(1):1-32.

33.

Galindo-Pumariño

Collado

Herrera

Pena

. Tumor microenvironment in metastatic colorectal cancer: The arbitrator in patients’ outcome. Cancers. 2021; Mar 6;13(5):1130.

34.

Al-Saleh

Abd El-Aziz

Ali

, et al. A multicenter study of the impact of body mass index (BMI) on the incidence of pathologic complete response (pCR) among Saudi patients with locally advanced breast cancer (LABC) post neoadjuvant chemotherapy (NAC). Gulf J Oncolog. 2019; May;1(30):33-42.

35.

Cabrera-Mulero

Crujeiras

Izquierdo

, et al. Novel SFRP2 DNA methylation profile following neoadjuvant therapy in colorectal cancer patients with different grades of BMI. J Clin Med. 2019; Jul 17;8(7):1041.

36.

Juszczyk

Kang

Putnis

, et al. High body mass index is associated with an increased overall survival in rectal cancer. J Gastrointest Oncol. 2020; Aug;11(4):626-632.

37.

De Nardi

Salandini

Chiari

, et al. Changes in body composition during neoadjuvant therapy can affect prognosis in rectal cancer patients: An exploratory study. Curr Probl Cancer. 2020; Apr;44(2):100510.

38.

Motoori

Fujitani

Sugimura

, et al. Skeletal muscle loss during neoadjuvant chemotherapy is an independent risk factor for postoperative infectious complications in patients with advanced esophageal cancer. Oncology. 2018;95(5):281-287.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.22 MB