Preoperative Sarcopenia is Associated with Poorer Compliance Rate of Adjuvant Chemotherapy and Worse Disease-free Survival of Patients with Stage (cid:0) Colon Cancer

Purpose: Preoperative sarcopenia has been proved to be associated with worse postoperative outcomes in cancer patients. This study aimed to evaluate whether preoperative sarcopenia affects the perioperative outcomes, adjuvant chemotherapy, and long-term outcomes of patients with stage (cid:0) colon cancer. Methods: Total 218 patients who underwent curative resection for stage (cid:0) colon cancer in our department from January, 2015 and December, 2018 were retrospectively analyzed. Sarcopenia was assessed by total psoas index, which measured the total area the level of L3 vertebral body and normalized according to patients’ height. Perioperative complications, postoperative adjuvant chemotherapy, and long-term prognosis were retrospectively analyzed. Results: Of 218 patients, 100(45.9%) patients were diagnosed with sarcopenia. Sarcopenia did not add the risk of perioperative complications (20.0% vs 15.3%, P=0.357), but it increased hospital stays (7.6±3.9 vs 6.7±2.2 days, P=0.042). Patients with sarcopenia had a lower rate of receiving adjuvant chemotherapy (70.0% vs 82.2%, p=0.033) and less likely to receive adequate adjuvant chemotherapy (58.6% vs 70.1, P=0.08). Patients with adequate adjuvant chemotherapy had signicantly better 3-year OS (89.8% vs 79.5, P=0.005) and a tendency of better 3-year DFS (76.4% vs 63.6%, P=0.055) than those with inadequate adjuvant chemotherapy and Non-adjuvant chemotherapy. Compared with the patients without sarcopenia, patients with sarcopenia had signicantly worse 3-year DFS (76.9% vs 62.8%, P=0.026). Conclusion: Preoperative sarcopenia was an important indicator to predict the compliance of AC in stage (cid:0) colon cancer patients, and it is also a signicant prognostic factor of worse 3-year DFS.


Introduction
Sarcopenia is de ned as a progressive and generalized skeletal muscle disorder that is associated with an increased likelihood of adverse outcomes including falls, fractures, physical disability, and mortality by the European Working Group on Sarcopenia in Older People (EWGSOP) in its latest guideline. [1] Sarcopenia can be exacerbated by the hypercatabolic state and in ammatory response caused by malignancy. [2] Therefore, Sarcopenia has been proposed to be an alternative marker of frailty for cancer patients in some study groups, [3][4][5] and preoperative sarcopenia is associated with poor postoperative outcomes, such as wound infection, anastomotic leakage, length of hospital stays, in patients with gastrointestinal surgery. [4,6] There also were some studies nding sarcopenia was a negative prognostic factor for disease-free survival (DFS) and overall survival (OS) in colorectal cancer patients [7,8] .
Sarcopenia is commonly diagnosed by measuring cross-sectional skeletal muscle area at the level of the third lumbar(L3) on the preoperative CT scan in most studies. [4,5,9,10] However, this method is often timeconsuming and complicated, and is not suitable for clinical application. Some studies tried to use the total psoas index (TPI), which measured the total area at the level of L3 vertebral body and normalized according to patients' height, to assess sarcopenia and satisfactory results were obtained [8,11,12] . The TPI is easy to measure and reproduced in clinical practice.
The association between TPI and surgical outcomes has been explored in recent years, and the results indicated it was a poor prognostic predictor in colorectal cancer [11] . However, whether low TPI (sarcopenia) affected the compliance of patients to receive the recommended standard adjuvant chemotherapy (AC) has not been well explored. According to the guideline of the National Comprehensive Cancer Network (NCCN), patients with stage colon cancer were recommended to receive AC because the results of multiple studies that demonstrated a clear bene ts of AC, [13][14][15][16] and the 6 months of oxaliplatin combined with either infusional 5-uorouacil plus folinic acid (FOLFOX) or capecitabine (CAPOX) became standard adjuvant regimen because of a consistent, albeit moderate, disease-free survival bene t compared with that of uorouracil plus folinic acid alone. [15,17] However, the compliance rate of AC for patients with stage colon cancer was unsatisfactory due to the complications, frailty after surgery, or economic reasons. Even in MOSAIC trial, only 74.7% patients completed the standard. [13] Sarcopenia was associated with weakness and delayed recovery after surgery, which may be one important risk factor of the compliance of AC. The present study aimed to evaluate whether sarcopenia affected the compliance of patients to receive the recommended standard AC and its impact on the long-term prognosis of stage colon cancer.

Study population
We retrospectively analyzed 1352 consecutive patients with localized colon cancer who underwent radical resection in the Department of Colorectal Surgery of Fujian Medical University Union Hospital (FMUUH) from January 2015 to December 2018. Patients who met the following criteria were included in the nal analysis: (1) pathologically con rmed as stage colon cancer according to the AJCC TNM staging system; (2) received abdominal CT scan within 30 days before surgery;

Treatment and follow-up
All the included patients underwent radical resection and were recommended to receive AC according to the guideline of the National Comprehensive Cancer Network (NCCN) [18] and the Ministry of Health of the People's Republic of China [19] . For p-Stage III colon cancer patients in our institute, the CAPOX (capecitabine and oxaliplatin) was a regular regime for AC. Adequate AC was de ned as receiving CAPOX for more than 3 months. Inadequate AC included patients that received CAPOX less than 3 months and that only received oral capecitabine. The follow-up evaluations were performed every 3 months for the rst 2 years, then every 6 months for the next 3 years, and annually thereafter. The postoperative followup included a physical examination, serum carcinoembryonic antigen (CEA) test, chest-to-pelvic CT, and colonoscopy. The baseline clinicopathological characteristics, postoperative complications (such as intestinal obstruction, anastomotic leak, abdominal infection, wound infection, pneumonia), recurrence status, and metastasis information of patients were collected from their medical records or through a telephone follow-up.

Image analysis
Total psoas area (TPA, mm 2 ) was retrospectively measured on the preoperative CT scan at the level of the third lumbar vertebral (L3). The exact level of measurement was de ned as the axial plane CT slice in which both L3 transverse processes were maximally in the view according to the earlier study. TPA was measured by manual outlining of both the left and right psoas muscle borders at the L3 level, and the area was automatically calculated on Picture Archiving and Communication System (PACS) software ( Fig 2). The area was then normalized by the square of the height (m 2 ) to obtain the total psoas index (TPI, mm 2 /m 2 ). The TPA was measured by two trained investigators (LY and GC) separately and the average was taken to nal analysis, both were blinded to the patient outcomes at the time of quanti cation. Sarcopenia was de ned using sex-speci c thresholds of <524 mm 2 /m 2 for male patients and <385 mm 2 /m 2 for female patients according to earlier studies [8] . Sarcopenia was de ned as an absolute variable and patients were classi ed as sarcopenia or non-sarcopenia in this study.

Statistical analysis
Categorical variables were present as frequencies and percentages, and Continuous variables were described as mean with standard deviation (SD) or interquartile range (IQR). Chi-square or Fisher's exact test was used to analyze categorical variables and a two-sample t-test or the Wilcoxon signed-rank test was performed to analyze continuous variables. The Kaplan-Meier method and the log-rank test were used for survival analysis. Multivariate analysis was conducted using a Cox proportional hazards model.
All analyses were performed with SPSS 23.0 for MAC (SPSS Inc., Chicago, IL, USA). A two-tailed P value <0.05 was considered statistically signi cant.

Results
Baseline demographic and clinicopathological characteristics

Adjuvant chemotherapy and survival analysis
There were 76.6% of total patients receiving AC, and 58.1% of them were non-sarcopenia. The rate of receiving AC in the non-sarcopenia group was signi cantly better than the sarcopenia group (82.2% vs 70%, p=0.033, Figure 3A). Patients without sarcopenia had more likely to receive adequate AC (more than 3 months) than patients with sarcopenia (70.1% vs 58.6, P=0.084, Figure 3B), though the difference did not reach signi cance. It should be noticed that patients who received AC had signi cantly better 3-year OS (87.6% vs 76.8%, P=0.007, Figure 4A) and DFS (73.6% vs 59.5%, P=0.034, Figure 4B) than patients without AC. In the subgroup analysis, when compared with patients who did not receive AC or received inadequate AC, patients who received adequate AC had signi cantly better 3-year OS (89.8% vs 79.5, P=0.005, Figure 4C) and a tendency of better 3-year DFS (76.4% vs 63.6%, P=0.055, Figure 4D).

Discussion
This study showed that sarcopenia, which was de ned by TPI, was signi cantly associated with a longer postoperative hospital stay in patients with stage colon cancer who underwent laparoscopic radical resection, and patients with sarcopenia had a poorer completion rate of adequate AC. Sarcopenia was also a signi cant prognostic predictor for 3-year DFS.
Nowadays, there are two main methods to quantify sarcopenia, including measuring cross-sectional skeletal muscle area or cross-sectional area of psoas muscles at the level of the L3. [20] In this study, we de ned sarcopenia by measuring a sectional area of psoas muscles due to its operability and clinical practicality.
Sarcopenia was diagnosed in 19.6% to 32.9% of colorectal cancer patients in early studies. [8,11] There were 45.9% of patients being classi ed as sarcopenia in our study. This may because we only included patients with stage colon cancer, which was more likely to cause frailty. We found patients with sarcopenia had longer postoperative hospital stay than those without sarcopenia, though the overall rate of postoperative morbidity was not signi cantly different between the two groups. A longer postoperative hospital stay may be due to the frailty caused by sarcopenia, which made the patients need more time to recover from the surgical trauma.
In the past 30 years, AC has been the standard treatment for patients with stage colon cancer since the previous studies have shown the bene ts in relapse-free survival and overall survival by using AC. [21,22] Consistent with early studies, our study showed AC signi cantly improved 3-year OS and DFS. However, due to economic, physical, or psychological factors, not all patients with stage colon cancer can nish the formal AC in clinical practice. [23] Roger H et al found that up to 40% of patients with stage or highrisk stage colon cancer did not receive recommended AC in their study. [24] According to the nal results of a prospective, pooled analysis of six randomized, phase 3 trials (IDEA collaboration), the 5-year overall survival between 3 months and 6months of AC was similar (82.1% versus 81.2%, HR 0.96 [0.85-1.08]), which supported the use of 3 months of adjuvant CAPOX for most patients with stage III colon cancer. [21] In our study, we de ne that receiving AC more than 3 months of CAPOX as an adequate AC. For the rst time, we found sarcopenia was a signi cant factor to in uence the compliance of receiving both AC and adequate AC for patients with stage colon cancer. Patients with sarcopenia had signi cantly lower compliance than those without sarcopenia. In addition, among patients who received AC, the rate of receiving adequate AC in the non-sarcopenia group was higher than in the sarcopenia group.
The reason for a lower rate of adequate AC in the sarcopenia group may be that the surgery trauma aggravated the frailty of patients who had sarcopenia, which makes them unable to withstand the toxicity and complications of AC. Nowadays, some studies explored the feasibility and e cacy of neoadjuvant chemotherapy for locally advanced colon cancer and demonstrated the potential bene t when compared to AC [25][26][27] . Considering the low compliance of AC in patients with sarcopenia, neoadjuvant chemotherapy may be an option, and better nutrition support for the malnourished sarcopenia patients before the operation may bene t the short-term and long-term outcome, [28] which need further study to con rm.
The prognostic signi cance of sarcopenia in patients with colorectal cancer (CRC) is controversial. A retrospective study of 220 stage -CRC found that sarcopenia was an independent risk factor for both recurrence-free survival (RFS) and OS. [6] However, another study including 494 patients with CRC showed sarcopenia did not signi cantly correlate with OS or RFS. [29] The reason for the different results between the two studies may be that they used different diagnostic criteria. In this study, when using the TPI as the diagnostic criteria for sarcopenia, we found sarcopenia was a signi cantly negative prognosis factor of 3-year DFS in patients with stage colon cancer. There was a tendency that patients with sarcopenia had poorer 3-year OS than those without sarcopenia (79.6% vs 90.0%, P=0.092), though the P-value did not reach the signi cance, the possible reason for these results is the relatively small sample size and inadequate follow up.
There were some limitations of this study that should be addressed. First, this was a single-institution, retrospective study with relatively small sample size. Second, we used the prede ned cutoff for sarcopenia in the previous study, which may be not appropriated for the population included in this study.
The optimal cut-off values for TPI that de ne sarcopenia in Asian patients with colon cancer, still require further investigation. Third, limited by data, the dose intensity of chemotherapy drugs was not analyzed in this study, and whether the dose intensity in uenced the completion rate of AC was not discussed.

Conclusion
To the best of our knowledge, this is the rst study to explore whether sarcopenia affects the compliance of patients to receive the recommended AC. Sarcopenia was an important indicator to predict the compliance of postoperative AC of stage colon cancer patients, and patients without sarcopenia had better compliance to the adjuvant chemotherapy. Sarcopenia was also a signi cant prognostic factor, which was associated with a worse 3-year DFS. To improve the compliance of AC and prognosis, it may be helpful to provide nutrition support and relieve sarcopenia before surgery, and further study on this topic is warranted. If possible, the more appropriate AC regimen should be administered according to the individual condition, and it can also further explore the feasibility of neoadjuvant chemotherapy in patients with sarcopenia.
Declarations Figure 1 Flow chart. Psoas muscle (red area) at the level of L3 vertebral body.