Use of Neoadjuvant Chemotherapy in the Treatment of Locally Advanced Rectal Cancer

j o u r n a l o f s u r g i c a l r e s e a r c h  n o v e m b e r 2 0 1 9 ( 2 4 3 ) 4 4 7 e4 5 2

Available online at www.sciencedirect.com

ScienceDirect journal homepage: www.JournalofSurgicalResearch.com

Association for Academic Surgery

Use of Neoadjuvant Chemotherapy in the Treatment of Locally Advanced Rectal Cancer Katherine Y. Hu, MD,a,1 Matthew T. Simpson, BS,a,1 Jacqueline J. Blank, MD,a Aniko Szabo, PhD,b Daniel Eastwood, MS,b Kirk A. Ludwig, MD,a Carrie Y. Peterson, MD, MS,a and Timothy J. Ridolfi, MDa,* a b

Division of Colorectal Surgery, Department of Surgery, Medical College of Wisconsin, Milwaukee, Wisconsin Division of Biostatistics, Medical College of Wisconsin, Milwaukee, Wisconsin

article info

abstract

Article history:

Background: Current treatment for locally advanced rectal cancer includes neoadjuvant

Received 1 March 2019

chemoradiation followed by surgery and adjuvant chemotherapy. With neoadjuvant

Received in revised form

chemotherapy (NC), both chemoradiation and chemotherapy are given in the neoadjuvant

9 May 2019

setting. This study aims to assess patterns of NC utilization and differences in treatment

Accepted 20 June 2019

response compared with standard treatment at our institution.

Available online xxx

Materials and methods: We performed a retrospective review of patients treated for stage IIIII rectal cancer at our institution between 2008 and 2018, examining patient demographics,

Keywords:

tumor characteristics, and treatment modality. The primary outcome of interest was

Rectal cancer

complete response (CR) to treatment, including both pathologic and clinical CR.

Chemotherapy

Results: Of 184 patients, 134 (72.8%) received standard therapy, and 50 (27.2%) received NC.

Rectal adenocarcinoma

In the standard treatment group, 70.1% were node positive, and 9.0% had T4-disease,

Neoadjuvant chemotherapy

compared with 92.0% and 26.0% in the NC group, respectively (both P < 0.01). NC utilization

Locally advanced rectal cancer

increased over time, with 3.4% of patients receiving NC between 2008 and 2012, compared with 48.5% in 2013-2018 (P < 0.01). CR was achieved in 19.4% versus 34.0% (P < 0.01) of patients in standard versus NC groups. With multivariate analysis, NC (odds ratio ¼ 3.02 [95% confidence interval 1.37-6.67], P ¼ 0.01) was associated with increased likelihood of achieving CR, whereas higher T-stage was associated with decreased likelihood of CR (for cT4, odds ratio ¼ 0.06 [95% confidence interval 0.01-0.56], P ¼ 0.01). Conclusions: Use of NC was increasingly used at our institution from 2008 to 2018. Patients who received NC achieved higher rates of CR compared with those undergoing standard therapy, despite having more advanced disease. These data support trends from other institutions and provides rationale for further study regarding use of NC for locally advanced rectal cancer. ª 2019 Elsevier Inc. All rights reserved.

Quickshot presentation at the 14th Annual Academic Surgical Congress, February 5-7, 2019, Houston, TX. * Corresponding author. Department of Surgery, Medical College of Wisconsin, 8701 Watertown Plank Road, Milwaukee, WI 53226. Tel.: þ1 414 955 5783; fax: þ1 414 955 0087. E-mail address: [email protected] (T.J. Ridolfi). 1 These authors contributed equally to this work. 0022-4804/$ e see front matter ª 2019 Elsevier Inc. All rights reserved. https://doi.org/10.1016/j.jss.2019.06.089

448

j o u r n a l o f s u r g i c a l r e s e a r c h  n o v e m b e r 2 0 1 9 ( 2 4 3 ) 4 4 7 e4 5 2

Introduction

Materials and methods

Current treatment of locally advanced (stage II or III) rectal adenocarcinoma consists of neoadjuvant chemoradiation, followed by total mesorectal excision and adjuvant chemotherapy.1 With standard treatment, approximately 8%-24% of patients are able to achieve a pathologic complete response (pCR), which has been associated with decreased rates of local recurrence and improved survival.2,3 One possible means to increase CR rates is to use neoadjuvant chemotherapy (NC), where systemic chemotherapy is given upfront in the preoperative setting, in addition to chemoradiation. A clinical trial conducted by Chau et al. was the first to explore outcomes associated with the use of NC. Patients with high-risk tumors, as determined by magnetic resonance imaging criteria, received neoadjuvant oxaliplatin and capecitabine before chemoradiation, achieving a 24% rate of pCR.4 Since then, several other phase II clinical trials have reported positive outcomes with NC regimens. In a multi-institutional study evaluating administration of mFOLFOX6 as consolidation chemotherapy administered after chemoradiation and before surgery, a higher proportion of patients with locally advanced disease (38%) were able to achieve pCR with a total neoadjuvant therapy (TNT) approach.5 A similar multicenter trial conducted by Marco et al. evaluated outcomes using mFOLFOX6 consolidation chemotherapy in addition to standard adjuvant chemotherapy. In their study, addition of NC increased rates of both pCR and disease-free survival compared with standard treatment.6 Induction chemotherapy regimens, where systemic chemotherapy is administered before chemoradiation, have also demonstrated improvement treatment response. A single-institution study of patients at Memorial Sloan Kettering Cancer Center showed that induction FOLFOX chemotherapy yielded a 36% CR rate and was well-tolerated with no serious adverse events.7 There are other possible benefits of NC beyond potential to improve rates of pCR. There may be improved delivery of chemotherapy to the primary tumor if given before vascular disruption caused by surgery and radiation.7 Total mesorectal excision for rectal cancer is associated with significant morbidity that can delay or prohibit adjuvant treatment. In one study, adherence to preoperative chemotherapy was 82.0% versus 42.9% postoperatively.8 Giving chemotherapy upfront before surgery may improve tolerance and completion rates of chemotherapy.1,9 For carefully selected patients, NC may potentially allow for non-operative management of rectal cancer. In addition, completing all planned chemotherapy upfront may reduce the time patients need a diverting ostomy.9 Although the exact role of NC in treatment of rectal cancer is still under evaluation, TNT is becoming more common and is now considered by the National Comprehensive Cancer Network (NCCN) as an acceptable treatment strategy in locally advanced rectal cancer (LARC).1,9 The aim of this study was to assess patterns of NC utilization at our institution, comparing differences in patient characteristics and treatment response in patients receiving standard treatment versus NC.

This was a retrospective cohort study of patients with LARC who received treatment at our institution, a 600-bed tertiarycare academic hospital, from January 1, 2008, through June 6, 2018. The study protocol was approved by the Medical College of Wisconsin Institutional Review Board with a waiver of informed consent. Data were collected at the patient level by querying the electronic medical record. LARC was defined as a clinical stage II or III (cT3/T4 N0 or any cT N1/N2) rectal adenocarcinoma, staged, and diagnosed per NCCN guidelines.1 Patients were excluded if they were referred for treatment of recurrent disease, had metastatic disease, or did not receive chemoradiation (i.e., underwent surgery alone or chemotherapy alone). Standard treatment was defined as neoadjuvant chemoradiation followed by planned total mesorectal excision and adjuvant chemotherapy. NC included any type of chemotherapy regimen administered in the neoadjuvant setting, including induction chemotherapy given before chemoradiation, consolidation chemotherapy given after chemoradiation but before surgery, and “sandwich” chemotherapy given in both the neoadjuvant and adjuvant settings. Those receiving only chemosensitizing doses during radiation therapy were not included in the NC group. There was no restriction by type or duration of NC regimen. We also included patients who completed their neoadjuvant treatment at their local hospital, as long as they underwent surgical resection and long-term follow-up at our institution. pCR was defined as no viable tumor cells in the resection specimen. Clinical CR (cCR) was defined as no evidence of disease on follow-up endoscopic or imaging evaluation. Patients were offered the option to defer surgery and “watch and wait” if they achieved cCR. We used the term CR in this study to identify patients who were documented to have either a pCR or cCR. Demographic data were analyzed with descriptive statistics. Treatment groups were compared using chi-square and Fisher’s exact tests for categorical variables and Wilcoxon rank-sum or KruskaleWallis tests for continuous variables. Multivariate logistic regression with stepwise model selection was used to test for possible significant multivariate variables to estimate the probability of achieving CR, including age, body mass index (BMI), gender, ethnicity, clinical T-stage, and use of oral anticoagulation, metformin, statins, NC, and adjuvant chemotherapy. All statistical analyses were performed by the Department of Biostatistics at the Medical College of Wisconsin using SAS version 9.4 (The SAS Institute, Cary, NC). Statistical significance was defined as a P-value <0.05.

Results Of the 184 patients with LARC who met inclusion criteria, 134 (72.8%) received standard treatment with CRT, and 50 (27.2%) received NC. There were no significant differences in age, sex, race, or BMI between the two groups. There was increased

hu et al  neoadjuvant chemotherapy in rectal cancer

institutional utilization of NC over time, with three of 87 patients (3.4%) receiving NC in 2008-2012, compared with 47 of 97 (48.5%) in 2013-2018 (P < 0.01). Frequency of NC utilization by year is illustrated in Figure. The NC group included a greater proportion of patients with T4 tumors (26.0% versus 9.0%, P < 0.01), node-positive disease (92.0% versus 70.1%, P < 0.01), and clinical stage III disease (92.0% versus 71.6%, P < 0.01) compared with the standard treatment group. The combined CR rate in the standard treatment group was 19.4% (n ¼ 26, with 7 cCR and 19 pCR) compared with 34.0% (n ¼ 17, with 10 cCR and 7 pCR) in the NC group. In the standard treatment group, 18.4% of patients with clinical stage II disease and 19.8% of patients with clinical stage III disease achieved CR. In contrast, all patients in the NC group who achieved CR had clinical stage III disease (Table 1). Nearly all patients underwent long-course neoadjuvant chemoradiation, with only two patients in the standard group treated with a short-course regimen. There was wide variability in total radiation dose, ranging from <4500 cGy to >5040 cGy, with most patients in both the standard and NC groups receiving 5040 cGy (60.4% and 82.0%, respectively). The chemosensitizing agent most commonly used in both groups was capecitabine (59.7% in the standard group and 78.0% in NC group). The most common NC regimen was mFOLFOX6based, with 64.0% of patients receiving this regimen. Of the 50 patients who received NC, 35 (70.0%) underwent TNT, whereas the remaining 15 patients (30.0%) also received adjuvant chemotherapy (Table 2). Multivariate logistic regression analysis identified NC to be significantly associated with achievement of CR (odds ratio [OR] 3.02, 95% confidence interval [CI] 1.37-6.67, P ¼ 0.01). Using cT1 disease as a reference, higher clinical Tstage was also associated with decreased likelihood of achieving CR (cT4 OR 0.06, 95% CI 0.01-0.56, P ¼ 0.01), although the relationship between cT2-3 disease and CR (OR 0.38, 95% CI 0.05-2.96, P ¼ 0.32; and OR 0.14, 95% CI 0.02-0.87, P ¼ 0.09, respectively) did not achieve statistical significance (Table 3).

449

Discussion This study revealed several patterns in utilization of NC at our institution over the past 10 y. Use of NC in patients with LARC has significantly increased over time, from only 3.4% of patients in the first half of our study period (2008-2012) to 48.5% in 2013-2018. This is consistent with the timing of new updates regarding NC in the literature, with many of the larger studies investigating NC published during the early to mid-2010s, as well as the update to the NCCN guidelines categorizing TNT as an acceptable treatment pathway for LARC. Despite typically having more advanced disease, a greater proportion of patients at our institution who received NC were able to achieve CR compared with the cohort who received standard treatment regimens, with 34% of patients achieving either a pCR or cCR. The association between NC and CR remained significant after multivariate analysis, with an associated OR of 3.02. In our analysis, clinical T-stage was the only other clinical factor significantly associated with CR, with higher T-stage associated with lower likelihood of achieving CR. Odds of achieving CR in patients with cT4 disease was 0.06 in comparison to those with cT1 tumors. Although our findings are limited to our own institution, they are consistent with trends reported by other institutions in the literature. A similar single-institution retrospective review of outcomes with TNT was performed at Memorial Sloan Kettering Cancer Center. In their study, 35.7% patients who received FOLFOX induction chemotherapy had either pCR or cCR, compared with 21.3% in the standard chemoradiation group. Similar to our findings, clinical T-stage and NC were significantly associated with CR. They additionally identified age >75 y and more recent timing of diagnosis to be associated with CR.9 Despite the growing body of evidence supporting its use, it is still unknown if NC is the optimal treatment for LARC and what type of regimen should be used. To date, the available evidence supporting the use of NC comes from retrospective

Fig e Frequency of institutional utilization of NC over time. Use of NC in patients with LARC has significantly increased over time at our institution, from only 3.4% of patients in 2008-2012 to 48.5% in 2013-2018.

450

j o u r n a l o f s u r g i c a l r e s e a r c h  n o v e m b e r 2 0 1 9 ( 2 4 3 ) 4 4 7 e4 5 2

Table 1 e Patient characteristics and treatment outcomes. Patient characteristics

Standard treatment (n ¼ 134)

NC (n ¼ 50)

Age, y Mean (SD) Median

0.16 57.8 (13) 56.0

60.2 (13.4) 61.0

Sex, n (%)

0.99

Male

83 (61.9)

31 (62.0)

Female

51 (38.0)

19 (38.0)

Race/ethnicity, n (%)

0.84

White/Caucasian

112 (83.6)

African American

10 (7.5)

4 (8.0)

Other

12 (9.0)

3 (6.0)

29.2 (7.0)

28.7 (5.7)

BMI, mean (SD)

43 (86.0)

2008-2012

84 (62.7)

3 (6.0)

2013-2018

50 (37.3)

47 (94.0) <0.01

cT classification, n (%) 1

5 (3.7)

2

14 (10.4)

3

100 (74.6)

32 (64.0)

4

12 (9.0)

13 (26.0)

3 (2.2)

2 (4.0)

Missing

0 3 (6.0)

<0.01

cN classification, n (%) 0

34 (25.4)

4 (8.0)

1

81 (60.4)

30 (60.0)

2

13 (9.7)

Missing

6 (4.5)

16 (32.0) 0 <0.01

Clinical stage, n (%) 2

38 (28.4)

4 (8.0)

3

96 (71.6)

46 (92.0)

0

19 (14.2)

7 (14.0)

1

35 (26.1)

6 (12.0)

2

23 (17.2)

10 (20.0)

3

44 (32.8)

4

2 (1.5)

No operation

9 (6.7)

Missing

2 (1.5)

<0.01

Pathologic stage, n (%)

14 (28.0) 0 13 (26.0) 0 <0.01

Underwent planned surgery, n (%) 125 (93.3)

37 (74.0)

Nodoffered watch/wait

7 (5.2)

10 (20.0)

Nodother*

2 (1.5)

3 (6.0)

CR, n (%)

26 (19.4)

17 (34.0)

pCR

19 (14.2)

7 (14.0)

cCR

*

0.95 <0.01

Time period, n (%)

Yes

P value

7 (5.2)

<0.01

10 (20.0)

Clinical stage II

7 of n ¼ 38 (18.4)

0

Clinical stage III

19 of n ¼ 96 (19.8)

17 of n ¼ 17 (100%)

Patient declined surgery or was no longer a surgical candidate because of comorbidities or metastatic disease.

review and phase II clinical studies. In addition, treatment regimens have varied in both interval and duration of treatment, making it unclear which chemotherapy regimen is

most effective. There has also been a wide array of chemotherapeutic combinations used, including induction and consolidation FOLFOX, induction capecitabine and

hu et al  neoadjuvant chemotherapy in rectal cancer

Table 2 e Details of neoadjuvant treatment regimens. Treatment

Standard treatment (n ¼ 134 [%])

NC (n ¼ 50 [%])

Total dose neoadjuvant radiation (cGy) Short coursed2140 over five fractions

1 (0.7)

0

Short coursed2500 over five fractions

1 (0.7)

0

<4500

5 (3.7)

2 (4.0)

4500-5039

9 (6.7)

1 (2.0)

5040

81 (60.4)

41 (82.0)

>5040

13 (9.7)

2 (4.0)

Missing

24 (17.9)

4 (8.0)

Chemoradiationdchemosensitizing agent Capecitabine Capecitabine/oxaliplatin 5-FU

80 (59.7)

0

29 (21.6)

6 (12.0)

None (radiation only) Missing NCdtype

39 (78.0)

7 (5.2)

0

1 (2.0)

18 (13.4)

4 (8.0)

N/A

mFOLFOX6

32 (64.0)

FOLFIRINOX

1 (2.0)

Capox/XELOX

6 (12.0)

Other FOLFOX*

8 (16.0)

Missing NCdtiming

3 (6.0) N/A

TNT

35 (70.0)

“Sandwich” chemotherapyy

15 (30.0)

* Initiated on FOLFOX-based chemotherapy, later combined with or replaced by other agents (bevacizumab, FOLFIRINOX, FOLFOXIRI) before course completion. y Sandwich chemotherapy included patients who received chemotherapy in both the neoadjuvant and adjuvant setting.

oxaliplatin, and induction bevacizumab plus oxaliplatin and capecitabine.4,6,7,10,11 This variability in the literature is reflected in our data, where multiple NC regimens were used during the study period. In our institution, all new rectal cancer cases are presented and discussed at a multidisciplinary tumor board, where a recommendation is made regarding whether or not a

451

patient would benefit from NC. Multiple variables are considered in this discussion, including performance status, comorbidities, patient goals, clinical stage, concerning imaging findings, and the feasibility of a sphincter-sparing operation. In addition to this initial discussion, if there are concerns or changes as a patient completes their neoadjuvant treatment (either NC or chemoradiation), they are re-presented and rediscussed at tumor board. If a consensus to recommend NC is reached, patients may either complete their course of NC at our institution or locally if it is more convenient. We do not currently have a standardized institutional NC protocol, which may contribute further to the variability in NC regimens seen in our data. Several ongoing randomized prospective trials may provide answers regarding which treatment regimen is most effective. The RAPIDO trial is a randomized phase III trial comparing short-course chemoradiation followed by selective adjuvant chemotherapy versus a TNT approach.12 The KONCLUDE trial will compare standard treatment consisting of neoadjuvant chemoradiation, surgery, and adjuvant chemotherapy against the same regimen but with the addition of neoadjuvant consolidation chemotherapy.13 Further study evaluating long-term overall and disease-free survival with NC is also needed. Limitations to our study include its retrospective design and small sample size. Our definition of CR was broad, as we did not restrict this by duration of sustained cCR. This may have confounded our results by potentially categorizing patients as having cCR when they may eventually develop a late recurrence. Because of our small sample size, we included all types of NC regimens in the NC cohort, without restriction on pharmacologic agents or timing (induction versus consolidation chemotherapy, TNT versus a “sandwich” approach with both neoadjuvant and adjuvant treatment, and so forth). Although this more inclusively captured use of NC in our institution, it added an additional confounding factor to our results.

Conclusions In our institution, NC for treatment of LARC was associated with higher rates of CR to treatment, despite its utilization in patients with more advanced disease. Regression analysis demonstrated NC, and lower clinical T-stage were independently associated with CR. Longer follow-up and further study are warranted to determine whether NC improves long-term outcomes and survival, and if so, which regimen is most effective.

Table 3 e Multivariate analysis of clinical variables associated with achieving CR. Variable

OR (95% CI)

P value

3.02 (1.37-6.67)

0.01

1

Reference

d

2

0.38 (0.05-2.96)

0.32

3

0.14 (0.02-0.87)

0.09

4

0.06 (0.01-0.56)

0.01

NC cT classification

Acknowledgment This research did not receive any specific grant funding from funding agencies in the public, commercial, or not-for-profit sectors. Authors’ contributions: T.J.R. was responsible for conception, design, and oversight of the project. M.T.S. and J.J.B. helped in data acquisition. M.T.S., A.S., D.E., K.Y.H., and T.J.R.

452

j o u r n a l o f s u r g i c a l r e s e a r c h  n o v e m b e r 2 0 1 9 ( 2 4 3 ) 4 4 7 e4 5 2

performed analysis and interpretation of the data. K.Y.H. and M.T.S. were responsible for the drafting of the article. K.A.L., C.Y.P., and T.J.R. performed critical review of the article.

6.

Disclosure 7.

The authors report no proprietary or commercial interest in any product or concept mentioned in this article. 8.

references 9. 1. National Comprehensive Cancer Network. NCCN Clinical Practice Guidelines - rectal cancer (version 3.2018). 2018. Available at: https://www.nccn.org/professionals/physician_ gls/PDF/rectal.pdf. Accessed November 29, 2018. 2. Maas M, Nelemans PJ, Valentini V, et al. Long-term outcome in patients with a pathological complete response after chemoradiation for rectal cancer: a pooled analysis of individual patient data. Lancet Oncol. 2010;11:835e844. 3. Garcı´a-Aguilar J, Hernandez de Anda E, Sirivongs P, Lee S-H, Madoff RD, Rothenberger DA. A pathologic complete response to preoperative chemoradiation is associated with lower local recurrence and improved survival in rectal cancer patients treated by mesorectal excision. Dis Colon Rectum. 2003;46:298e304. 4. Chau I, Brown G, Cunningham D, et al. Neoadjuvant capecitabine and oxaliplatin followed by synchronous chemoradiation and total mesorectal excision in magnetic resonance imagingedefined poor-risk rectal cancer. J Clin Oncol. 2006;24:668e674. 5. Garcia-Aguilar J, Chow OS, Smith DD, et al. Effect of adding mFOLFOX6 after neoadjuvant chemoradiation in locally

10.

11.

12.

13.

advanced rectal cancer: a multicentre, phase 2 trial. Lancet Oncol. 2015;16:957e966. Marco MR, Oommen S, Stamos MJ, et al. Consolidation mFOLFOX6 chemotherapy after Chemoradiotherapy improves survival in patients with locally advanced rectal cancer. Dis Colon Rectum. 2018;61:1146e1155. Cercek A, Goodman KA, Hajj C, et al. Neoadjuvant chemotherapy first, followed by chemoradiation and then surgery, in the management of locally advanced rectal cancer. J Natl Compr Canc Netw. 2014;12:513e519. Bosset J-F, Daban A, Calais G, et al. Chemotherapy with preoperative radiotherapy in rectal cancer. N Engl J Med. 2006;355:1114e1123. Cercek A, Roxburgh CSD, Strombom P, et al. Adoption of total neoadjuvant therapy for locally advanced rectal cancer. JAMA Oncol. 2018;4:e180071. Gao Y-H, Lin J-Z, An X, et al. Neoadjuvant sandwich treatment with oxaliplatin and capecitabine administered prior to, concurrently with, and following radiation therapy in locally advanced rectal cancer: a prospective phase 2 trial. Int J Radiat Oncol. 2014;90:1153e1160. Nogue M, Salud A, Vicente P, et al. Addition of bevacizumab to XELOX induction therapy plus concomitant capecitabinebased chemoradiotherapy in magnetic resonance imagingdefined poor-prognosis locally advanced rectal cancer: the AVACROSS study. Oncologist. 2011;16:614e620. Nilsson PJ, van Etten B, Hospers GAP, et al. Short-course radiotherapy followed by neo-adjuvant chemotherapy in locally advanced rectal cancer–the RAPIDO trial. BMC Cancer. 2013;13:279. Kim CW, Kang BM, Kim IY, et al. Korean Society of Coloproctology (KSCP) trial of cONsolidation Chemotherapy for Locally advanced mid or low rectal cancer after neoadjUvant concurrent chemoraDiothErapy: a multicenter, randomized controlled trial (KONCLUDE). BMC Cancer. 2018;18:538.