Original Article. Division of Gastrointestinal and General Surgery, Department of Surgery, 2 Department of Radiation Medicine, 3

Transcription

1 Diseases of the Esophagus (2017) 30, 1 7 DOI: /dote/dox015 Original Article Neoadjuvant chemoradiotherapy with concurrent cisplatin/5-fluorouracil is associated with increased pathologic complete response and improved survival compared to carboplatin/paclitaxel in patients with locally advanced esophageal cancer K. R. Haisley, 1 K. D. Hart, 1 N. Nabavizadeh, 2 K. G. Bensch, 3 G. M. Vaccaro, 3 C. R. Thomas, Jr, 2 P. H Schipper, 4 J. G Hunter, 1 J. P. Dolan 1 1 Division of Gastrointestinal and General Surgery, Department of Surgery, 2 Department of Radiation Medicine, 3 Division of Medical Oncology, Department of Internal Medicine, and 4 Division of Cardiovascular and General Thoracic Surgery, Department of Surgery, Oregon Health and Science University, Portland, OR, USA SUMMARY. Trimodal therapy consisting of neoadjuvant chemoradiation followed by esophagectomy has become the standard of care in North America for locally advanced esophageal cancer. While cisplatin/5-fluorouracil has been a common concurrent chemotherapy regimen since the 1980s, its utilization has declined in recent years as the Chemoradiotherapy for Oesophageal Cancer Followed by Surgery Study (CROSS) trial regimen of carboplatin/paclitaxel has become widely adopted. The efficacy of the CROSS regimen compared to alternate chemotherapy choices, however, has rarely been evaluated when each is used as a component of a trimodal treatment approach. The aim of this study is to report our institutional experience with these two concurrent chemotherapy regimens at a specialized esophageal cancer center. We performed an Institutional Review Board-approved retrospective review of a prospectively maintained institutional foregut registry from a single National Cancer Institute-designated cancer center. Esophageal cancer patients who completed trimodal therapy with a chemotherapy regimen of either carboplatin/paclitaxel or cisplatin/5- fluorouracil were identified and divided into groups based on their chemotherapy regimens. Multivariable logistic regression was used to analyze pathologic complete response rates, while the Kaplan Meier and Cox proportional hazards models were utilized to evaluate recurrence-free and overall survival. Analytical models were adjusted for age, clinical stage, radiation dose, histologic subtype (adenocarcinoma vs. squamous cell carcinoma), and time interval from completion of neoadjuvant therapy to surgery. One hundred and forty-two patients treated between January of 2000 and July of 2015 were identified as meeting inclusion criteria. Of this group, 87 had received the CROSS regimen of carboplatin/paclitaxel, while 55 had completed cisplatin/5-fluorouracil. Multivariable analysis demonstrated that the cisplatin/5-fluorouracil.group had an increased odds of pathologic complete response (odds ratio = 2.68, 95% confidence interval, P = 0.032), as well as significantly improved recurrence-free survival (hazard ratio = 0.39, 95% confidence interval , P = 0.003) and overall survival (hazard ratio = 0.46, 95% confidence interval , P = 0.016), compared to the carboplatin/paclitaxel group. Concurrent chemotherapy with cisplatin/5-fluorouracil in locally advanced esophageal cancer is associated with higher rates of pathologic complete response and improved recurrence-free and overall survival compared to the CROSS regimen of carboplatin/paclitaxel. This suggests that, for select patients, alternate neoadjuvant chemotherapy approaches, such as cisplatin/5-fluorouracil, merit reconsideration as potential primary treatment choices in the management of this highly morbid disease. KEY WORDS:: carboplatin + paclitaxel, cisplatin + 5-fluorouracil, esophageal cancer, neoadjuvant chemotherapy, pathologic complete response. Address correspondence to: James Dolan, MD, MCR, 3181 SW Sam Jackson Park Road, Mailcode L223A, Portland, OR 97239, USA. dolanj@ohsu.edu Conflicts of interest: The authors declare that they have no conflict of interest. Funding Sources: None. C The Authors Published by Oxford University Press on behalf of International Society for Diseases of the Esophagus. All rights reserved. For permissions, please journals.permissions@oup.com 1

2 2 Diseases of the Esophagus INTRODUCTION Esophageal cancer is an increasingly prevalent disease that accounts for greater than 15,000 deaths in the United States every year. 1 Even early-stage cancers have been associated with five-year survival rates as low as 37%, with more advanced presentations having survival times measured in months. 1 The high morbidity and mortality associated with the diagnosis of esophageal cancer highlights the need to refine treatment strategies in order to better optimize outcomes. In recent years, some progress has been made with the addition of concurrent chemoradiation to the standard treatment protocol for esophageal cancers. This trimodal treatment approach, consisting of neoadjuvant chemotherapy and radiation followed by esophagectomy, has gained considerable favor since the publication of the Chemoradiotherapy for Oesophageal Cancer Followed by Surgery Study (CROSS) in This landmark trial demonstrated a markedly improved five-year survival rate in patients receiving trimodal therapy versus those having surgery alone. As a result, trimodal therapy has become the standard of care in North America for locally advanced esophageal cancer. To date, the efficacy of the CROSS chemotherapy regimen of carboplatin/paclitaxel has not been directly compared to alternate chemotherapy regimens when each is used as a component of a full trimodal treatment approach. 3 Our working hypothesis was that outcomes would not differ between the CROSS regimen and the conventional regimen utilized prior to this study. Hence, the primary aim of this study is to analyze our institutional experience, comparing the CROSS regimen of carboplatin/paclitaxel to a conventional regimen of cisplatin/5-fluorouracil, including only patients who underwent a trimodal treatment approach. METHODS Study design and participants We performed a retrospective review of a prospectively maintained esophageal disease registry from a single National Cancer Institute-designated cancer center under Institutional Review Board-approved (IRB#15198). We selected patients who had completed neoadjuvant chemoradiotherapy (CRT) followed by esophagectomy for documented esophageal cancer between January of 2000 and July of Preoperative tumor staging was performed using endoscopy, computed tomography (CT) scanning, and radiolabeled fluorodeoxyglucose whole body positron emission tomography. Each patient s case was discussed in detail at a biweekly esophageal care conference. Decisions regarding the type of chemotherapy administered were left to the discretion of the treating oncologist. Those patients who had received a chemotherapy regimen of either cisplatin/5-fluorouracil or carboplatin/paclitaxel were included in the study. Patients who received any other chemotherapy regimen, or who had missing or incomplete treatment data were excluded from further analysis. Patients were then divided into two groups based on their administered chemotherapy regimens, either carboplatin/paclitaxel or cisplatin/fluorouracil. Statistical analysis Demographic and clinical characteristics of patients included in the analysis were analyzed using descriptive statistics. Analytical models were adjusted for age, clinical stage, radiation dose, histologic subtype (adenocarcinoma vs. squamous cell carcinoma), and time interval from completion of neoadjuvant therapy to surgery. We utilized multivariable logistic regression to analyze pathologic complete response rates and the Kaplan Meier plots, as well as Cox proportional hazards models to analyze recurrence-free and overall survival designating the day of surgery as time zero. RESULTS Patient demographics Of the 529 patients in our esophageal diseases registry at the time of our query, 248 had completed trimodal therapy with a diagnosis of esophageal cancer. Twenty-six patients were excluded due to our inability to determine what specific chemotherapy regimen was administered. Seventy additional patients were removed from analysis due to having received any one of 27 unique chemotherapy regimens that were neither of the regimens under current review. These regimens ranged from single agent to quadruple agent, and were largely platinum based. No one of these alternate regimens was administered in more than eight cases. An additional 10 patients had other missing data variables that required their exclusion from the study. In total, 142 patients were identified who had undergone neoadjuvant CRT with one of the two regimens of interest. Fifty-five patients received cisplatin/5-fluorouracil, and the remaining 87 were given carboplatin/paclitaxel (Fig. 1). There was no significant difference between groups in terms of ethnicity, gender, tumor histology, or adjuvant therapies received (Table 1). In the carboplatin/paclitaxel group, the mean age at surgery was more advanced (65.6 vs. 60.4, P = 0.003), and median time between completion of neoadjuvant therapy and surgery was longer (64 vs. 56 days, P = 0.001). The cisplatin/5-fluorouracil group appeared to show a trend toward having higher stage cancers, though this did not reach statistical significance (P = 0.07).

3 Neoadjuvant CRT in esophageal cancer 3 overall in the cisplatin/5-fluorouracil group with a complete response rate of 33% compared to a rate of only 22% in the carboplatin/paclitaxel group, though this finding did not reach statistical significance on univariable analysis (P = 0.21). Under multivariable analysis correcting for age, histologic subtype, clinical stage, neoadjuvant radiation dose, and days between CRT and surgical resection, the cisplatin/5- fluorouracil group had a significantly increased odds of pathologic complete response (odds ratio 2.67, 95% confidence interval: , P = 0.032, Table 2). Fig. 1 Patient enrollment flowsheet. Pathologic complete response Overall, 37 patients (26%) achieved a pathologic complete response as determined by final pathology. There was a trend toward increased pathologic complete Survival On the Kaplan Meier analysis, median recurrencefree survival time (including recurrence at any site, both locoregional or metastatic) was higher among patients who received cisplatin/5-fluorouracil (median: 2.5 years) than among those who received carboplatin/paclitaxel (median: 1.25 years). Recurrence-free survival probabilities at one, two, and three years were higher in the cisplatin/5- fluorouracil group (70%, 53%, and 44%) than in Table 1 Demographic information for our patient cohort All patients Cisplatin/5-FU Carboplatin/paclitaxel P-value (n = 142) (n = 55) (n = 87) Age ± ± ± Gender Male 82% (117) 80% (44) 84% (73) Female 18% (25) 20% (11) 16% (14) Ethnicity Caucasian 98% (139) 96% (53) 99% (86) American Indian 1% (1) 0% (0) 1% (1) Asian 1% (1) 2% (1) 0% (0) Other 1% (1) 2% (1) 0% (0) Histologic subtype Adenocarcinoma 86% (122) 85% (47) 86% (75) SCC 14% (20) 15% (8) 14% (12) Clinical stage 1 2% (3) 2% (1) 2% (2) % (43) 27% (15) 33% (28) 3 62% (88) 60% (33) 64% (55) 4 5% (7) 11% (6) 1% (1) Radiation dose (cgy) 5040 (5000, 5040) 5040 (4770, 5040) 5040 (5000, 5040) Adjuvant chemotherapy Yes 5% (6) 4% (2) 5% (4) No 95% (124) 96% (52) 95% (72) Days from CRT to surgery 62 (52.25, 73.00) 56 (48.00, 69.00) 64 (56.00, 85.00) Pathologic node positivity No 58% (83) 69% (38) 52% (45) Yes 42% (59) 31% (17) 48% (42) Pathologic complete response Yes 26% (37) 33% (18) 22% (19) No 74% (105) 67% (37) 78% (68) Student s t-test (two-sided); Pearson s chi-squared test; Fisher s exact test; Wilcoxon rank-sum test with continuity correction. FU, fluorouracil; SCC, squamous cell carcinoma; x:x (x:x, x:x) indicates median and inter-quartile range; x±x indicates mean ± standard deviation.

4 4 Diseases of the Esophagus Table 2 Multivariable logistic regression exploring pathologic complete response Clinical variable OR (95% CI) P-value Cisplatin/5-fluorouracil (vs. carboplatin/paclitaxel) (1.086, 6.610) Age (per year) (0.980, 1.076) Stage III of IV (vs. I or II) (0.172,0.954) Squamous cell carcinoma (vs. adenocarcinoma) (0.550, 5.127) Days from ncrt to surgery (0.999, 1.001) Total radiation dose (0.999, 1.001) ncrt, neoadjuvant chemoradiotherapy. Fig. 2 Kaplan Meyer plot of (a) recurrence-free and (b) overall survival as a factor of chemotherapy regimen (shaded areas denote 95% confidence interval). the carboplatin/paclitaxel group (56%, 31%, and 31%) (Fig. 2a). Similarly, median overall survival time was higher among patients who received cisplatin/5-fluorouracil (6.8 years) than among those who received carboplatin/paclitaxel (1.8 years). Further, predicted overall survival probabilities at one, two, and three years were higher in the cisplatin/5-fluorouracil group (76%, 60%, and 52%) than in the carboplatin/paclitaxel group (69%, 36%, and 24%) (Fig. 2b).

5 Neoadjuvant CRT in esophageal cancer 5 Table 3 Multivariable Cox proportional Hazards model of overall and recurrence-free survival Clinical variable Overall survival Recurrence-free survival (n = 133) (n = 133) HR (95% CI) HR (95% CI) cisplatin/5-fluorouracil (vs. carboplatin/paclitaxel) (0.244,0.865) (0.209,0.727) Age (per year) (0.964,1.022) (0.269,1.497) Stage III of IV (vs. I or II) (0.935,3.323) (1.270,4.314) Squamous cell carcinoma (vs. adeno) (0.238,1.557) (0.269, 1.497) Days from ncrt to surgery (0.980,1.007) (0.977, 1.003) Total radiation dose (0.999,1.000) (0.999,1.000) P < 0.1. P < P < In a multivariable analysis, patients who received a neoadjuvant chemotherapy regimen of cisplatin/5- fluorouracil had significantly improved recurrencefree (hazard ratio: 0.39, 95% confidence interval: , P = 0.003) and overall survival (hazard ratio: 0.46, 95% confidence interval , P = 0.016), compared to the carboplatin/paclitaxel group (Table 3). DISCUSSION In this study, we have shown that concurrent chemotherapy with cisplatin/5-fluorouracil is associated with higher rates of pathologic complete response and improved recurrence-free and overall survival, compared to the widely utilized CROSS regimen of carboplatin/paclitaxel when each is used as a component of trimodal therapy for locally advanced esophageal cancer. The adoption of standardized trimodal therapy has resulted in significant improvements compared to surgery alone. 1 2,4 However, the question of which patients will most benefit from CRT 5 and which neoadjuvant chemotherapy regimen might be best for optimizing outcomes and survival remains unanswered. 6 While cisplatin/5-fluorouracil was previously a commonly utilized regimen, the CROSS regimen of carboplatin/paclitaxel has largely become the standard of therapy over the last five years. It is important to recognize that the use of carboplatin/paclitaxel in the CROSS trial, and its subsequent adoption nationwide, may have been more a reflection of institutional preference and tolerability, rather than a direct comparison of this regimen versus any other in terms of survival or oncologic outcomes. Since the publication of the CROSS trial and the rapid spread of the carboplatin/paclitaxel concurrent chemotherapy regimen, several authors have retrospectively evaluated its efficacy compared to alternate chemotherapy choices. A recent metaanalysis concluded that carboplatin/paclitaxel was the superior regimen, 3 while other authors have shown no difference between the two regimens in terms of survival. 7,8 Still others have shown finding similar to our study and suggest that carboplatin/paclitaxel might be associated with higher recurrence 9 and lower rates of pathologic complete response 10 than cisplatin/5-fluorouracil. Taken together, while each of these studies has certain limitations, the cumulative literature highlights the need for ongoing research in this field. The improved survival rates we and other authors have shown following completion of neoadjuvant CRT with cisplatin/5-fluorouracil certainly has biologic plausibility. Our findings may be a result of the increased rates of pathologic complete response (elimination of all detectable tumor from the surgical specimen) seen with this more aggressive regimen. 7,8,11 Several studies have demonstrated that pathologic complete response in esophageal cancer is associated with improved survival While the exact factors that predict or increase the likelihood of pathologic complete response remain an area of intense research, 13 these data support the idea that the choice of chemotherapy may significantly affect rates of response. As cisplatin/5-fluorouracil appears to be associated with higher rates of pathologic complete response in this study, it stands to reason that it may eliminate tumor more effectively than carboplatin/paclitaxel, thus reducing recurrences and improving survival. Comparing our results to those of the CROSS trial highlights the finding that our institution did not find equivalent survival rates to those published in the CROSS trial when using the regimen of carboplatin/paclitaxel (49 months in the CROSS versus only 21.7 months in our population). This may be due to our smaller sample size compared to the CROSS group as well as a larger proportion of squamous histology seen in the CROSS cohort. In addition, it is fairly well established that adenocarcinoma patients (86% of our patients vs. 75% of the CROSS patients) have lower rates of pcr and shorter median survivals. 2 In addition, our higher median age at surgery (65.6 years) compared to the CROSS patients (60 years) may have contributed to increased perioperative mortality. Finally, the higher doses of

6 6 Diseases of the Esophagus radiation used at our institution (50.4 Gy) compared to the published CROSS regimen of (41.4 Gy) may also contribute to increased toxicity and decreased survival. 17 It is worth mentioning that the tolerability profile of carboplatin/paclitaxel is generally considered to be superior to that of cisplatin/5-fluorouracil in terms of side effects and patient experience. While direct dosing and tolerability data could not be expressly collected in this study due to the disparate locations across a wide geographical area where our patients received their chemotherapy, other authors have demonstrated this to be the case. 7,8 It is also important to recognize that the better tolerability of the carboplatin/paclitaxel regimen may create a manner of selection bias such that weaker patients with more comorbidities are making it through neoadjuvant therapy who would have otherwise failed on the cisplatin/5fu regimen. Nonetheless, treatment decisions must be individualized, based on the patient s risk for complications from chemotherapy versus their risk of cancer spread or recurrence. In the era of personalized medicine, it will be interesting to address if stage, age, or other clinical factors might predict which patients would most benefit from one regimen over the other. More research will be required to determine definitively which patients might benefit from a more intensive chemotherapy regimen such as cisplatin/5- fluorouracil, versus those that will do well with a more tolerable regimen of carboplatin/paclitaxel. We recognize that there are a number of other limitations to this study. First, it is retrospective in nature. However, our data were conducted prospectively under a single institution with standardized protocols for staging, treatment, and specimen analysis. We are further limited by our relatively small sample size, with only 142 patients meeting criteria for inclusion. In addition, there is also almost certainly a temporal effect within our data, as the cisplatin/5-fluorouracil group was treated mostly between the years of 2000 and 2012, while the carboplatin/paclitaxel group was treated more recently. There have been other care pathway changes for our patients between these time periods that cannot be fully accounted for in our statistical modeling. This might include improved ICU care, implementation of our institutional esophagectomy pathway, and evolution of surgical technique and practice. Furthermore, intensity-modulated radiation therapy (IMRT) has seen increased utilization over-time, concurrent with carboplatin/paclitaxel use. 14 It is well known that thoracic irradiation can cause acute and chronic lung injury and absolute volume of the total lung spared from radiation doses greater than 5 Gy has been strongly associated with risk of postoperative pulmonary complications. 15 Compared to traditional, three-dimensional conformal radiotherapy techniques, the modulation of multiple beams inherent to IMRT exposes larger proportions of the bilateral lung to a low-dose bath of radiation. Given that a very large proportion of patients receiving carboplatin/paclitaxel were treated with IMRT it is difficult to remove this confounding variable in our analysis. At our institution, the total radiation dose was maintained at 50.4 Gy both before and after the transition a carboplatin/paclitaxel chemotherapy regimen rather than adopting the CROSS dosing of 41.4 Gy. This gives consistency to the radiation dosing between eras. Despite recent advancements, esophageal cancer remains a prevalent and deadly malignancy with an urgent need for further refined treatment strategies. Future studies evaluating the efficacy of various neoadjuvant approaches, such as the currently enrolling PROTECT trial, 16 as well as evaluation of different radiation doses and modalities 17 will be an integral part of improving outcomes of this highly morbid disease. Such research is vital to further elucidate the role of alternate neoadjuvant chemotherapy regimens in esophageal cancer. CONCLUSION In our patient population, concurrent chemotherapy with cisplatin/5-fluorouracil in locally advanced esophageal cancer is associated with higher rates of pathologic complete response and improved recurrence-free and OS, compared to the CROSS regimen of carboplatin/paclitaxel. These data suggest that alternate neoadjuvant concurrent chemotherapy approaches merit reconsideration as potential primary treatment options in the management of this disease. ACKNOWLEDGEMENTS We would like to acknowledge the contributions of Charlie Borzy, our Study Coordinator, as well as Hope Hardaker, our Database Manager. We additionally thank Mary Kwatkosky-Lawlor for her invaluable assistance with manuscript preparation, editing, and formatting. References 1 Walsh T N, Grennell M, Mansoor S, Kelly A. Neoadjuvant treatment of advanced stage esophageal adenocarcinoma increases survival. Dis Esophagus 2002; 15: Van Hagen P, Hulshof M, Van Lanschot J et al. Preoperative chemotherapy for esophageal or junctional cancer. N Engl J Med 2012; 366: Huang T C, Hsu C H, Lin C C, Tu Y K. Systematic review and network meta-analysis: neoadjuvant chemoradiotherapy for locoregional esophageal cancer. Jpn J Clin Oncol 2015; 45: Siddiqui F A, Atkins K M, Diggs B S, Thomas C R, Jr, Hunter J G, Dolan J P. Overall survival analysis of neoadjuvant

7 Neoadjuvant CRT in esophageal cancer 7 chemoradiotherapy and esophagectomy for esophageal cancer. J Gastrointest Oncol 2014; 5: Eil R, Diggs B S, Wang S J, Dolan J P, Hunter J G, Thomas C R. Nomogram for predicting the benefit of neoadjuvant chemoradiotherapy for patients with esophageal cancer: a SEER-Medicare analysis. Cancer 2014; 120: Adenis A, Mirabel X, Mariette C. Is preoperative chemoradiation with paclitaxel and carboplatin a new standard of treatment for esophageal cancer? Int J Radiat Oncol Biol Phys 2013; 86: Blom R L, Sosef M N, Nap M et al. Comparison of two neoadjuvant chemoradiotherapy regimens in patients with potentially curable esophageal carcinoma. Dis Esopahagus 2014; 27: Honing J, Smit J, Muijs C et al. A comparison of carboplatin with paclitaxel and cisplatinum with 5-fluorouracil in definitive chemoradiotherapy in esophageal cancer patients. Ann Oncol 2014; 25 (3): Thomay A A, Su S, Friedant A J et al. Freedom from recurrence after induction cisplatin/5-fu/rt versus carboplatin/paclitaxel/rt in patients with esophageal cancer (Abstract 126). J Clin Oncol 2014; 32 ( Suppl 3): pp Schellenberg D, Peixoto R, Lee A, Lim H J. A province-wide retrospective comparison of pathologic response and progressionfree survival in locally advanced esophagus cancer patients treated with either cisplatin-5fu(cf) or carboplatin-paclitaxel (CP) concurrent with radiation (XRT) with or without surgery (Abstract 107). J Clin Oncol 2014; 32( Suppl 3): pp Berger A C, Farma J, Scott W J et al. Complete response to neoadjuvant chemoradiotherapy in esophageal carcinoma is associated with significantly improved survival. J Clin Oncol 2005; 23: Donahue J M, Nichols F C, Li Z et al. Complete pathologic response after neoadjuvant chemoradiotherapy for esophageal cancer is associated with enhanced survival. Ann Thorac Surg 2009; 87: LinSH,WangJ,AllenPKet al. A nomogram that predicts pathologic complete response to neoadjuvant chemoradiation also predicts survival outcomes after definitive chemoradiation for esophageal cancer. J Gastrointest Oncol 2015; 6: Simpson D R, Lawson J D, Nath S K, Rose B S, Mundt A J, Mell L K. A survey of image-guided radiation therapy use in the United States. Cancer 2010; 116: Wang S L, Liao Z, Vaporciyan A A et al. Investigation of clinical and dosimetric factors associated with postoperative pulmonary complications in esophageal cancer patients treated with concurrent chemoradiotherapy followed by surgery. Int J Radiat Oncol Biol Phys 2006; 64: Messager M, Mirabel X, Tresch E et al. Preoperative chemoradiation with paclitaxel-carboplatin or with fluorouraciloxaliplatin-folinic acid (FOLFOX) for resectable esophageal and junctional cancer: the PROTECT-1402, randomized phase 2 trial. BMC Cancer 2016; 16: Nabavizadeh N, Shukla R, Elliott D A et al. Preoperative carboplatin and paclitaxel-based chemoradiotherapy for esophageal carcinoma: results of a modified CROSS regimen utilizing radiation doses greater than 41.4 Gy. Dis Esophagus 2016; 29: