Risk Factors and Adequate Management for Complications of Bevacizumab Treatment Requiring Surgical Intervention in Patients With Metastatic Colorectal Cancer

Accepted Manuscript Risk factors and adequate management for complications of bevacizumab treatment requiring surgical intervention in patients with metastatic colorectal cancer Jun Woo Bong, MD, Jong Lyul Lee, MD, PhD, Chan Wook Kim, MD, PhD, Yong Sik Yoon, MD, PhD, In Ja Park, MD, PhD, Seok-Byung Lim, MD, PhD, Chang Sik Yu, MD, PhD, Tae Won Kim, MD, PhD, Jin Cheon Kim, MD, PhD PII:

S1533-0028(18)30265-2

DOI:

10.1016/j.clcc.2018.06.005

Reference:

CLCC 478

To appear in:

Clinical Colorectal Cancer

Received Date: 28 May 2018 Revised Date:

23 June 2018

Accepted Date: 25 June 2018

Please cite this article as: Bong JW, Lee JL, Kim CW, Yoon YS, Park IJ, Lim S-B, Yu CS, Kim TW, Kim JC, Risk factors and adequate management for complications of bevacizumab treatment requiring surgical intervention in patients with metastatic colorectal cancer, Clinical Colorectal Cancer (2018), doi: 10.1016/j.clcc.2018.06.005. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT Original Article Risk factors and adequate management for complications of bevacizumab treatment

RI PT

requiring surgical intervention in patients with metastatic colorectal cancer

Running title: Bevacizumab complications

Jun Woo Bong, MD1; Jong Lyul Lee, MD, PhD1; Chan Wook Kim, MD, PhD1; Yong Sik

SC

Yoon, MD, PhD1; In Ja Park, MD, PhD1; Seok-Byung Lim, MD, PhD1; Chang Sik Yu, MD,

1

M AN U

PhD1; Tae Won Kim MD, PhD2 and Jin Cheon Kim, MD, PhD1

Department of Surgery and Division of Colon and Rectal Surgery, Asan Medical Center,

University of Ulsan College of Medicine, Seoul, Korea 2

Department of Medical Oncology, Asan Medical Center, University of Ulsan College of

TE D

Medicine, Seoul, Korea

Corresponding author: Jin Cheon Kim, MD, PhD, Professor, Department of Surgery, Asan

EP

Medical Center, University of Ulsan College of Medicine, 88, Olympic-ro 43-gil, Songpa-gu, Seoul 05505, Korea

AC C

E-mail: [email protected] Tel: 82 2 3010 3489 Fax: 82 2 474 4028

Declarations of Interest: None.

1

ACCEPTED MANUSCRIPT Funding Sources: Grants (to J.C. Kim) from the Korea Research Foundation (2016R1E1A1A02919844), Ministry of Science, ICT, and Future Planning, Republic of

AC C

EP

TE D

M AN U

SC

RI PT

Korea

2

ACCEPTED MANUSCRIPT MicroAbstract:

The rate of complications requiring surgical intervention after bevacizumab

treatment for metastatic colorectal cancer was 5.9% in the current study. Not only the location or resected status of the primary tumor, but also the aggressiveness and stent inserted status

RI PT

of the primary tumor were identified as the risk factors. Bowel resection was performed in 50 percent but about 42% of complications were managed with performing diverting ostomy. Close post-treatment observation is necessary for patients prone to complications that require

AC C

EP

TE D

M AN U

SC

surgical intervention to maintain the survival time and sustain the treatment.

3

ACCEPTED MANUSCRIPT ABSTRACT Background: Bevacizumab has been approved for treating colorectal cancer since 2004. Although bevacizumab use may lead to adverse effects, few studies have reported incidences

RI PT

requiring surgical intervention. We aimed to identify the risk factors and adequate interventions for complications requiring surgical intervention following bevacizumab treatment.

SC

Patients and Methods: We retrospectively reviewed the records of patients with metastatic colorectal cancer treated with bevacizumab in our institute from January 2009 to December

complications requiring surgery.

M AN U

2016. The baseline patient characteristics were used to evaluate the potential risk factors for

Results: Of the 1008 patients recruited for this study, 60 (5.9%) experienced complications necessitating surgery after bevacizumab therapy. Gastrointestinal perforation was the most

TE D

frequently observed complication, occurring in 36 (3.5%) patients, and diverting colostomy was the most commonly performed intervention (22 patients; 36.6%). Multivariate analysis helped identify poor differentiation, signet-ring cell carcinoma, stent-insertion status, rectal

EP

location of the primary tumor, and intact primary tumor status as the risk factors. Survival time remained unchanged regardless of a complication that required surgery.

AC C

Conclusion: Careful monitoring during bevacizumab treatment for metastatic colorectal cancer is essential for patients who have a predisposition to complications that may require surgery. After detection, adequate and timely surgical management is imperative for ensuring patient survival.

Keywords: Bevacizumab; complications; metastatic colorectal cancer

4

ACCEPTED MANUSCRIPT 1. Introduction Bevacizumab (BV) is a recombinant humanized monoclonal antibody against vascular endothelial growth factor (VEGF). BV use was approved for treating metastatic colorectal

RI PT

cancer (mCRC) by the U.S. Food and Drug Administration in February 2004[1]. BV inhibits tumor growth by suppressing the growth of new blood vessels, reducing interstitial fluid pressure, and enhancing the efficacy of chemotherapeutic agents.[2] The substantial clinical

SC

efficacy of BV has been demonstrated in several studies—it prolongs overall and progression-free survival when used in combination with other chemotherapy regimens in the

M AN U

first- or second-line treatment of mCRC[3,4]. However, using BV is not free of adverse effects—hypertension and proteinuria have been reported as the most common adverse effects observed, and their rates of occurrence are 26%–37% and 10%–34.9%, respectively [3,5,6]

. Nonetheless, most complications are generally tolerable and manageable with standard

TE D

therapies, and BV is, therefore, accepted as an effective treatment for mCRC[3,4]. In rare cases, potentially life-threatening complications such as gastrointestinal perforation (GIP), severe bleeding, and arterial thromboembolism can also occur[5-7]. GIP, which is a representative

EP

complication requiring surgical intervention, occurs at a rate of 1%–2%[6,8]. In most cases, severely adverse events require immediate intervention for minimizing treatment delays and

AC C

disease progression. To date, however, only a few studies have examined the risk factors for complications requiring surgery after BV treatment owing to their low incidence. Although the incidence of complications requiring surgery is reportedly extremely low, the physicians at our institute have observed an increase in their frequency. In the current retrospective study, we describe the risk factors and management strategies for patients with complications that require surgery after BV treatment.

5

ACCEPTED MANUSCRIPT 2. Materials and Methods 2.1 Data collection We retrospectively reviewed the data of patients with mCRC who were treated with BV at

RI PT

our institute between January 2009 and December 2016. We included patients aged >18 years using an Eastern Cooperative Oncology Group performance status of 0–2 at the time of BV therapy commencement. Of the 1125 patients, 21 patients were excluded because they

SC

received chemotherapy treatments not including fluoropyrimidine-based regimens and BV. In addition, 32 patients who were previously treated for mCRC at another hospital, along with

M AN U

64 patients with a history of malignancies other than colorectal cancer, were also excluded. Thus, 1008 patients were included in our study, and their medical records were reviewed. Complications requiring surgery were defined as those requiring interventions under general anesthesia and having potentially life-threatening consequences without surgery. The

TE D

complications were confirmed via abdominopelvic computed tomography scans illustrating intraperitoneal air, the fistula tract, peritoneal abscess, and other clinical features of peritonitis. We categorized the patients into two groups according to the occurrence of complications

EP

requiring surgical intervention. In total, 948 and 60 patients were respectively categorized into complications and no complications groups (Figure 1). The Institutional Review Board

AC C

of Asan Medical Center approved this study (IRB approval number: S2017-1139-0001).

2.2 Treatment

Fluoropyrimidine-based chemotherapy regimens and tumor assessments were performed at a physician’s discretion at our institute. The BV dose was 5 mg/kg every 2 weeks for 5-FUbased regimens and 7.5 mg/kg every 3 weeks for capecitabine-based regimens; chemotherapy was discontinued upon disease progression or following patient refusal. The interval between 6

ACCEPTED MANUSCRIPT major surgery and BV was at least 30 days in all patients. When a complication requiring surgical intervention occurred, all interventions were carried out within 24 h of detection, and BV was permanently discontinued. BV was also permanently discontinued when arterial

RI PT

thromboembolism, proteinuria, or symptomatic venous thrombosis were detected.

2.3 Assessment

SC

The baseline patient characteristics such as age, sex, body weight, and body mass index (BMI) at the start of BV treatment were collected. The clinicopathological factors of the primary

M AN U

tumor such as the location, histological grades of differentiation, and histological subtypes were examined. Anticoagulant use and the presence of a stent in the large intestine during BV therapy were also determined. The presence of an intact primary tumor was ascertained prior to BV treatment. Other data such as radiation therapy history (including preoperative or

TE D

adjuvant radiation therapy), recent colonoscopy (or sigmoidoscopy) within 30 days before starting BV treatment, and whether BV was used in the first-line setting for mCRC were collected for analyzing the risk factors for complications requiring surgery. In addition,

EP

patients who experienced complications within 7 weeks of the last BV dose were studied to identify complications most likely caused by BV. We defined mortality as death within 30

AC C

days after the occurrence of complications requiring surgery. Overall survival (OS), which was defined as the time from the commencement of BV treatment to death or the date of last follow-up, was assessed for both the groups.

2.4 Statistical analysis Data were analyzed using SPSS version 21.0 (SPSS, Chicago, IL, USA). Hazard ratios and 95% confidence intervals (CIs) for risk factors were determined using Cox proportional 7

ACCEPTED MANUSCRIPT hazards regression analysis. Backward stepwise elimination was used for reducing potential confounding, and additional analyses were performed using a p-value of <0.05 as a criterion for covariate selection. Then, a two-sided p-value of <0.05 was considered as statistically

RI PT

significant. OS in both groups was analyzed using the Kaplan–Meier method and log-rank

AC C

EP

TE D

M AN U

SC

test.

8

ACCEPTED MANUSCRIPT 3. Results 3.1 Complications requiring surgery and their management The incidence rate of complications requiring surgery after BV treatment was approximately

RI PT

5.9% (60/1008) as per our retrospective analysis. GIP was the most common complication of this type (3.5%; 36/1008), and the most common site of perforation was the primary tumor in the colon or rectum (Table 1). Anastomosis dehiscence was the second-most common

SC

complication (1.5%; 16/1008). Diverting colostomy was the most frequently performed surgery to manage prior surgery-related complications (36.6%, 22/60, Table 2). Four patients

M AN U

died (6.6%), including two patients who refused surgery and died after conservative treatments—one died of hepatic failure following Hartmann’s operation, and the other died of septic shock after diversion colostomy. OS was not significantly different between the two groups (p = 0.585; Figure 2). The median OS was 19.4 months (95% CI = 15.4–23.3) in the

complication group.

TE D

surgery-related complication group, and 17.9 months (95% CI = 16.6–19.1) in the no

EP

3.2 Risk factors for complications requiring surgery We examined several baseline characteristics in the two groups (Table 3). The

results

of

AC C

univariate and multivariate analyses of the risk factors which required surgery were shown in the table 4. Multivariate analysis showed that the stent-insertion status, poorly differentiated carcinoma, signet-ring cell carcinoma, primary rectal cancer, and an intact primary tumor were significant risk factors for complications requiring surgery. Additionally, forty-five patients (75%) experienced complications requiring surgery within 7 weeks after the last BV dose. Multivariate analysis of these patients revealed that the complications requiring surgery were associated with the same aforementioned risk factors. However, for the patients who 9

ACCEPTED MANUSCRIPT experienced complications later than 7 weeks after the last BV dose, the same analysis

AC C

EP

TE D

M AN U

SC

RI PT

identified no significant risk factor for complications requiring surgery.

10

ACCEPTED MANUSCRIPT 4. Discussion Although self-expandable metal stents (SEMSs) permit a low-risk treatment of colonic obstruction either palliatively or preoperatively, they carry an overall complication rate of up

RI PT

to 25%[9,10]. Stent migration may occur at a rate of 5%–22% owing to tumor shrinkage following chemotherapy[11,12]. Tumor invasion into the lumen through the stent also causes colonic obstruction in 9%–10% of the cases, and perforation occurs in 7%–11% of the

SC

cases[12]. Our findings show that 7 patients experienced perforation at the primary tumor site after SEMS insertion and subsequent BV exposure. However, further evaluation is required to

M AN U

clarify whether the status of stent insertion during BV treatment is a real risk factor for GIP. According to other retrospective studies, BV increases the risk of perforation after SEMS insertion[12,13]. These studies reported an approximately three-fold higher rate of GIP in patients who underwent SEMS implantation and subsequently received BV than in those who

TE D

were not treated with BV after SEMS insertion. Several mechanisms have been proposed, including perforation due to BV-promoting tumor regression, necrosis, and a weakened serosa, combined with the pressure of the radial force of the stent on the colonic tumor[12,14].

EP

However, owing to the low number of occurrences and the retrospective nature of the studies, a precise risk of perforation has not yet been determined.

AC C

Poor differentiation and signet-ring cell carcinoma have been reported as risk factors for tumor aggressiveness[15-17]. Although the mechanism underlying this aggressiveness has not been reported, several studies have linked it to high levels of VEGF, which is the primary target of BV[18]. High VEGF levels result in an invasive colorectal cancer with a high degree of vascularity and a possibility of metastasis to distant tissues or organs[19]. BV plays a key role in preventing increased microvascular density and boosting the delivery of the baseline chemotherapeutic regimen to the tumor. Therefore, it is possible that the efficacy of BV may 11

ACCEPTED MANUSCRIPT be high in patients with poorly differentiated and/or signet-ring cell colorectal cancer; however, a high risk of ischemic change in the tumor because of the prevention of neovascularization must also be considered. Very few studies have explored this subject, and

differentiation, signet-ring cell carcinoma, and BV efficacy.

RI PT

additional research is required for clarifying the relationship between VEGF levels, poor

Although 30 patients (50%) underwent bowel resection owing to complications

SC

requiring surgery, 25 patients (41.6%) were diagnosed in the early phase of inflammation, and they underwent diversion ileostomy or colostomy without bowel resection. The mortality

M AN U

rate associated with surgery-requiring complications after BV use was 6.6% (4/60 patients). After excluding two patients who refused surgery, the median survival time after complications in the remaining 58 patients was 13.1 ± 4.9 months (95% CI = 3.5–22.7 months). Moreover, OS did not differ between the two groups. Therefore, if a complication

TE D

requiring surgery is identified, then immediate and active treatment is necessary for ensuring patient survival. In addition, clinicians should closely observe the status of patients with the identified risk factors for diagnosing and managing these complications as early as possible.

EP

Our current findings indicate that primary rectal cancer is a risk factor for surgical complications. Among the 33 complications requiring surgery in patients with primary rectal

AC C

cancer, 17 patients (51.5%) showed anastomosis-related complications after low anterior resection, including anastomosis dehiscence or rectovaginal fistula. High ligation of the inferior mesenteric artery was performed in approximately half of the patients. However, high ligation of the inferior mesenteric artery has also been reported as a risk factor for anastomotic complications[20]. Following high ligation of the inferior mesenteric artery, adequate vascular supply to the proximal colon after low anterior resection mostly depends on the patency of the marginal artery of Drummond from the middle colic artery[21]. In 12

ACCEPTED MANUSCRIPT addition, anatomical variations such as the absence of the middle colic artery or inadequate vascularization of the splenic flexure are frequent[22]. BV may worsen insufficient vascular supply at the anastomosis site by preventing the wound-healing process of the anastomoses.

RI PT

Because it impairs blood supply, radiotherapy is a possible primary cause of anastomotic complications. Our univariate analysis also shows that a history of radiotherapy is a risk factor. Although radiotherapy did not remain significant in the multivariate analysis, it has been reported as a significant risk factor for BV-associated bowel perforation even in the

SC

absence of anastomosis[8,23,24]. The deleterious effects of radiotherapy include inhibiting the

M AN U

wound-healing process and causing hypoxia at the site of tissue injury, including early vessel damage, angiogenesis inhibition, and fibroblast impairment[25]. With BV-associated vascular supply insufficiency, radiotherapy may worsen ischemic damage to the bowel wall and delay the healing process of anastomosis.

TE D

Previous studies have demonstrated that the location of the primary tumor is a predictive factor for the prognosis of mCRC [26,27]. Wong et al. showed that the survival outcome of right colon cancer was worse than that of left colon and rectal cancer but the

EP

survival benefit of BV was better in right colon cancer[26]. Other studies supported these results, suggesting that the genetic mutations (such as KRAS, BRAF and MSI-high) occur

AC C

more frequent in right-sided cancer and the VEGF expression, an important predictive factor for the response to BV appears lower in left-sided cancer (left colon and rectum) [28,29]. In our study, left-sided cancer represented tendency towards more complications requiring surgery in a univariate analysis (HR 2.041 95%CI 0.924-4.506, p = 0.078; table 4), possibly occurring from limited number of patients. A major limitation of our study was the possibility of unmeasured confounding factors because of the complicated mechanisms of neovascularization of the primary tumor and the 13

ACCEPTED MANUSCRIPT antiangiogenic effect of BV. In addition, the number of patients with complications requiring surgery was disproportionate to that of the patients without such complications because of their extremely low incidence. The fact that some of the risk factors analyzed in this study

RI PT

were consistent with those of previous studies might be another limitation of this study. Despite these limitations, this study may be a good guidance for physicians to help determine the treatment strategy for metastatic colorectal cancer patients in using bevacizumab through

AC C

EP

TE D

M AN U

SC

showing new risk factors and results of various treatments.

14

ACCEPTED MANUSCRIPT 5. Conclusion To summarize, the patients with risk factors for complications requiring surgery after BV treatment should be closely monitored for ensuring early detection of the complications.

RI PT

Upon identifying a complication, immediate intervention is essential for ensuring patient

AC C

EP

TE D

M AN U

SC

survival.

15

ACCEPTED MANUSCRIPT Clinical Practice Points
Several risk factors physicians should consider before using bevacizumab for metastatic colorectal cancer were identified in this study: the location, resection

RI PT

status, aggressiveness and stent insertion status of the primary tumor
These complications can seriously affect patients in a short period of time.


However, if the adequate treatment is performed on time, it may not worsen the

AC C

EP

TE D

M AN U

SC

survival of the patient.

16

ACCEPTED MANUSCRIPT References 1.

Hapani S, Chu D, Wu S: Risk of gastrointestinal perforation in patients with cancer treated with bevacizumab: a meta-analysis. The Lancet Oncology 2009;10:559-568.

2.

Jain RK, Tong RT, Munn LL: Effect of Vascular Normalization by Antiangiogenic Therapy on Mathematical Model. Cancer research 2007;67:2729-2735.

3.

RI PT

Interstitial Hypertension, Peritumor Edema, and Lymphatic Metastasis: Insights from a Giantonio BJ, Catalano PJ, Meropol NJ, et al.: Bevacizumab in combination with oxaliplatin, fluorouracil, and leucovorin (FOLFOX4) for previously treated metastatic colorectal cancer: results from the Eastern Cooperative Oncology Group Study E3200. J Clin Oncol 2007;25:1539-1544.

Hurwitz HI, Fehrenbacher L, Hainsworth JD, et al.: Bevacizumab in combination with

SC

4.

fluorouracil and leucovorin: an active regimen for first-line metastatic colorectal cancer. J

Clin Oncol 2005;23:3502-3508.

Van Cutsem E, Rivera F, Berry S, et al.: Safety and efficacy of first-line bevacizumab with

M AN U

5.

FOLFOX, XELOX, FOLFIRI and fluoropyrimidines in metastatic colorectal cancer: the BEAT study. Ann Oncol 2009;20:1842-1847. 6.

Kozloff M, Yood MU, Berlin J, et al.: Clinical outcomes associated with bevacizumabcontaining treatment of metastatic colorectal cancer: the BRiTE observational cohort study.

Oncologist 2009;14:862-870. 7.

Hurwitz HI, Saltz LB, Van Cutsem E, et al.: Venous thromboembolic events with

TE D

chemotherapy plus bevacizumab: a pooled analysis of patients in randomized phase II and III studies. J Clin Oncol 2011;29:1757-1764. 8.

Kabbinavar FF, Flynn PJ, Kozloff M, et al.: Gastrointestinal perforation associated with bevacizumab use in metastatic colorectal cancer: results from a large treatment

9.

EP

observational cohort study. Eur J Cancer 2012;48:1126-1132. Ptok H, Meyer F, Marusch F, et al.: Palliative stent implantation in the treatment of malignant colorectal obstruction. Surgical endoscopy 2006;20:909-914. Meisner S, Hensler M, Knop FK, et al.: Self-expanding metal stents for colonic obstruction:

AC C

10.

experiences from 104 procedures in a single center. Dis Colon Rectum 2004;47:444-450.

11.

Camunez F, Echenagusia A, Simo G, et al.: Malignant colorectal obstruction treated by

means of self-expanding metallic stents: effectiveness before surgery and in palliation.

Radiology 2000;216:492-497.

12.

Small AJ, Coelho-Prabhu N, Baron TH: Endoscopic placement of self-expandable metal stents for malignant colonic obstruction: long-term outcomes and complication factors.

Gastrointest Endosc 2010;71:560-572. 13.

Imbulgoda A, MacLean A, Heine J, et al.: Colonic perforation with intraluminal stents and bevacizumab in advanced colorectal cancer: retrospective case series and literature review. 17

ACCEPTED MANUSCRIPT Can J Surg 2015;58:167-171. 14.

Cennamo V, Fuccio L, Mutri V, et al.: Does stent placement for advanced colon cancer increase the risk of perforation during bevacizumab-based therapy? Clin Gastroenterol

Hepatol 2009;7:1174-1176. 15.

Barresi V, Reggiani Bonetti L, Ieni A, et al.: Poorly Differentiated Clusters: Clinical Impact in Colorectal Cancer. Clin Colorectal Cancer 2017;16:9-15. Nitsche U, Zimmermann A, Spath C, et al.: Mucinous and signet-ring cell colorectal

RI PT

16.

cancers differ from classical adenocarcinomas in tumor biology and prognosis. Ann Surg 2013;258:775-782; discussion 782-773. 17.

Schneider NI, Langner C: Prognostic stratification of colorectal cancer patients: current perspectives. Cancer Manag Res 2014;6:291-300.

Mohamed HAD AEAH, Kamel AAEA, et al: Correlation of Vascular Endothelial Growth

SC

18.

Factor Expression and Neovascularization with Colorectal Carcinoma: A Pilot Study. J 19.

M AN U

Adenocarcinoma 2016.

Wang Y, Yao X, Ge J, et al.: Can vascular endothelial growth factor and microvessel density be used as prognostic biomarkers for colorectal cancer? A systematic review and metaanalysis. ScientificWorldJournal 2014;2014:102736.

20.

Trencheva K, Morrissey KP, Wells M, et al.: Identifying important predictors for anastomotic leak after colon and rectal resection: prospective study on 616 patients. Ann Surg 2013;257:108-113.

Deshaies I, Malka D, Soria JC, et al.: Antiangiogenic agents and late anastomotic

TE D

21.

complications. Journal of surgical oncology 2010;101:180-183. 22.

Boni L, David G, Dionigi G, et al.: Indocyanine green-enhanced fluorescence to assess bowel

perfusion

2016;30:2736-2742.

laparoscopic

colorectal

resection.

Surgical

endoscopy

Badgwell BD, Camp ER, Feig B, et al.: Management of bevacizumab-associated bowel

EP

23.

during

perforation: a case series and review of the literature. Ann Oncol 2008;19:577-582. Sugrue M, Kozloff M, Hainsworth J, et al.: Risk factors for gastrointestinal perforations in

AC C

24.

patients with metastatic colorectal cancer receiving bevacizumab plus chemotherapy.

Journal of Clinical Oncology 2006;24:3535-3535.

25.

Vujaskovic Z, Anscher MS, Feng QF, et al.: Radiation-induced hypoxia may perpetuate late

normal tissue injury. International journal of radiation oncology, biology, physics

2001;50:851-855.

26.

Wong HL, Lee B, Field K, et al.: Impact of Primary Tumor Site on Bevacizumab Efficacy in Metastatic Colorectal Cancer. Clin Colorectal Cancer 2016;15:e9-e15.

27.

He WZ, Liao FX, Jiang C, et al.: Primary Tumor Location as a Predictive Factor for First-line Bevacizumab Effectiveness in Metastatic Colorectal Cancer Patients. J Cancer 2017;8:38818

ACCEPTED MANUSCRIPT 394. 28.

Stintzing S, Tejpar S, Gibbs P, et al.: Understanding the role of primary tumour localisation in colorectal cancer treatment and outcomes. Eur J Cancer 2017;84:69-80. Ulivi P, Scarpi E, Chiadini E, et al.: Right- vs. Left-Sided Metastatic Colorectal Cancer:

EP

TE D

M AN U

SC

RI PT

Differences in Tumor Biology and Bevacizumab Efficacy. Int J Mol Sci 2017;18.

AC C

29.

19

ACCEPTED MANUSCRIPT Tables Table 1. Categories of complications requiring surgery (60 patients) No. of patients (%)

Gastrointestinal tract perforation

34 (56.6)

Small intestine

11 (18.3)

Colon

4 (6.6)

Stomach

1 (1.6)

Unidentified

2 (3.3)

SC

18 (30.0)

M AN U

Site of perforation

Primary tumor

RI PT

Complications

Anastomosis dehiscence

16 (26.6)

Rectovaginal fistula

4 (6.6)

Surgical wound dehiscence

2 (3.3)

2 (3.3)

AC C

EP

TE D

Bleeding

20

ACCEPTED MANUSCRIPT Table 2. Management of complications requiring surgery (60 patients) No. of patients (%)

Without bowel resection

26 (43.3)

Diverting colostomy

22 (36.6)

Diverting ileostomy

3 (5.0)

Primary closure

1 (1.6)

With bowel resection

30 (50)

Resection and anastomosis of small bowel

9 (15)

Resection and anastomosis of colon

8 (13.3)

Other

4 (6.6)

Wound revision

2 (3.3)

2 (3.3)

AC C

EP

TE D

Conservative management

SC

13 (21.6)

M AN U

Hartmann’s operation

RI PT

Management

21

ACCEPTED MANUSCRIPT Table 3. Baseline patient characteristics Patients without surgery-requiring Patients with surgery-requiring

characteristics

complications, n = 948 (%)

complications, n = 60 (%)

Age, years a

57.16 ± 11.51

57.22 ± 11.65

Female

397 (41.8)

23 (38.3)

Male

551 (58.1)

37 (61.6)

Body weight, kg a

61.70 ± 11.14

62.20 ± 10.29

BMI, kg/m2 a

23.14 ± 3.29

23.08 ± 3.10

No

888 (93.6)

Yes

60 (6.3)

First-line BV exposure 374 (39.4)

Yes

574 (60.5)

53 (88.3) 7 (11.6)

10 (16.6)

TE D

No

M AN U

Anticoagulant use

SC

Sex

Stent insertion

932 (93.6)

EP

No Yes

16 (6.3)

50 (83.3)

53 (88.3) 7 (11.6)

AC C

Previous radiation therapy No

795 (83.8)

44 (73.3)

153 (16.1)

16 (26.6)

No

786 (82.9)

43 (71.6)

Yes

162 (17.0)

17 (28.3)

Yes

RI PT

Baseline

Recent colonoscopy

Intact primary tumor 22

ACCEPTED MANUSCRIPT No

703 (74.1)

35 (58.3)

Yes

245 (25.8)

25 (41.6)

Right colon

224(23.6)

7(11.7)

Left colon

323(34.1)

16(26.6)

Rectum

362 (38.1)

33 (55.0)

Double primary

39 (4.1)

4 (6.6)

SC

Histologic differentiation of the primary tumor 816 (86.0)

46 (76.6)

Poor

105 (11.0)

11 (18.3)

AC

895 (94.4)

MAC

41 (4.3)

SRCC

12 (1.2)

M AN U

Well or Moderate

Histologic subtype of the primary tumor

RI PT

Primary tumor location

53 (88.3) 3 (5.0)

3 (5.0)

TE D

a: mean ± standard deviation; AC, adenocarcinoma; BV, bevacizumab; MAC, mucinous

AC C

EP

adenocarcinoma; SRCC, signet-ring cell carcinoma

23

ACCEPTED MANUSCRIPT Table 4. Univariate and multivariate Cox proportional hazards regression analyses of the risk factors for complications requiring surgery Multivariate HR

95% CI

p-value

1.001 (0.979–1.023)

0.941

Sex 1.0

Male

1.156 (0.687–1.945)

0.586

Body weight, kg

1.003 (0.980–1.026)

0.793

BMI, kg/m2

0.992 (0.918–1.073)

Anticoagulant use No

1.0

Yes

1.800 (0.818–3.961)

No

1.0

Yes

2.825 (1.431–5.575)

0.144

EP

Stent insertion

1.0 0.003

1.0

1.816 (0.907–3.634)

0.092

1.0

7.248 (3.292–15.957)

<0.001

AC C

No Yes

0.848

TE D

First-line BV use

p-value

M AN U

Female

95% CI

RI PT

Age, years

Univariate HR

SC

Baseline characteristics

5.687 (2.372–13.637)

<0.001

Previous radiotherapy No Yes

1.0

1.0

1.831 (1.033–3.245)

0.038

1.773 (0.878–3.578)

0.110

Recent colonoscopy No

1.0

Yes

1.846 (1.053–3.237)

1.0 0.032 24

0.752 (0.369–1.529)

0.431

ACCEPTED MANUSCRIPT Intact primary tumor No

1.0

1

Yes

2.095 (1.238–3.547)

0.006

1.904 (1.061, 3.415)

0.031

Primary tumor location (1) 1

Rectum

2.136 (1.254–3.637)

0.005

Primary tumor location (2) Right colon

1

Left colon + Rectum

2.041 (0.924-4.506)

0.078

Well or Moderate

1.0

Poor

2.253 (1.164–4.360)

Histologic subtypes of the primary tumor

2.611 (1.486–4.586)

M AN U

Histologic differentiation of the primary tumor

RI PT

1.0

0.001

SC

Right + Left Colon

1

0.016

2.118 (1.064–4.215)

0.033

0.015

1

0.017

1

MAC

1.497 (0.542–4.137)

0.437

1.255 (0.445–3.537)

0.668

SRCC

5.450 (1.695–17.525)

0.004

5.523 (1.700–17.940)

0.005

TE D

AC

AC, adenocarcinoma; CI, confidence interval; HR, hazard ratio; MAC, mucinous adenocarcinoma; SRCC,

AC C

EP

signet-ring cell carcinoma

25

ACCEPTED MANUSCRIPT Figures legends

Figure 1. Flowchart illustrating patient selection. (BV = bevacizumab; mCRC = metastatic

RI PT

colorectal cancer)

AC C

EP

TE D

M AN U

SC

Figure 2. Overall survival of patients with and without complications that required surgery.

26

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT