ANTICANCER RESEARCH 34: 5037-5042 (2014)

Irinotecan plus in Patients with Extensive-Disease Poorly Differentiated Neuroendocrine Carcinoma of the Esophagus

HITOMI SUMIYOSHI OKUMA1,2, SATORU IWASA1, HIROKAZU SHOJI1, ATSUO TAKASHIMA1, NATSUKO OKITA1, YOSHITAKA HONMA1, KEN KATO1, TETSUYA HAMAGUCHI1, YASUHIDE YAMADA1 and YASUHIRO SHIMADA1

1Gastrointestinal Medical Oncology Division, National Cancer Center Hospital, Tokyo, Japan; 2Department of Medical Oncology, Graduate School of Medicine, Chiba University Hospital, Chiba, Japan

Abstract. Background: Poorly-differentiated neuroendocrine from grades 1 and 2 neuroendocrine tumors according to the carcinoma (NEC) of the esophagus is a rare subtype that has a 2010 World Health Organization classification (1). In Japan, poor prognosis and is distinguished from well-differentiated a nationwide survey performed in 2005 estimated the neuroendocrine neoplasms in accordance with the 2010 World incidence of GI neuroendocrine tumors as 2.10 per 100,000 Health Organization classification. -plus-cisplatin people (2). GI-NEC is estimated to account for 9.3% of all (IP) is used as first-line for extensive-disease GI neuroendocrine tumors (3). Among all esophageal cancers, (ED) small-cell and its use is plausible for first- the frequency of NEC ranges from 0.05% to 7.6% (4-8). It is line chemotherapy in ED esophageal NEC. We retrospectively a poorly-differentiated carcinoma with pathological features analyzed the efficacy and toxicity of IP for ED esophageal mimicking those of small-cell lung cancer (SCLC) and has a NEC. Patients and Methods: Patients with ED esophageal NEC poor prognosis (9). However, because of its rarity, no treated with IP between 2000 and 2013 were retrospectively standard-of-care with strong evidence has been established. identified from our database. The end-points were objective Patients with extrapulmonary NEC are usually treated response rate, progression-free survival (PFS) and overall with the same chemotherapy protocols as those used for survival (OS). Data on adverse events were also collected. SCLC. In a phase III evaluation of extensive-disease (ED)- Results: An objective response was achieved in 50% (95% SCLC, irinotecan-plus-cisplatin (IP) was compared against confidence interval [CI]: 25% to 75%) of 12 identified patients. -plus-cisplatin (EP), and IP provided improved Median progression-free survival was 4.0 months (95% CI: 0.9 progression-free survival (PFS) and overall survival (OS) to 7.6) and overall survival was 12.6 months (95% CI: 4.6 to (10). Therefore, IP has been adopted as the standard-of-care 28.6). Grade 3/4 hematological toxicities included leukopenia in Japan for patients with ED-SCLC. Herein, we in 50% of patients and neutropenia in 67%. The rate of febrile retrospectively assessed the efficacy and toxicity of IP for neutropenia was 25%. No treatment-related deaths were ED esophageal NEC. observed. Conclusion: IP appears acceptable as first-line chemotherapy for ED esophageal NEC. Patients and Methods

Neuroendocrine carcinoma (NEC) of the gastrointestinal (GI) Database. Patients who had been treated between January 2000 and tract is a rare histological subtype classified as poorly- April 2013 at the National Cancer Center Hospital (Tokyo, Japan) were identified from patient databases. The study was approved by differentiated neuroendocrine carcinoma and distinguished the Institutional review board.

Data extraction. A retrospective review was performed to evaluate the outcomes in 12 cases of primary ED and recurrent Correspondence to: Iwasa Satoru, MD, Gastrointestinal Medical neuroendocrine carcinoma of the esophagus treated with IP. The Oncology Division, National Cancer Center Hospital, 5-1-1 Tsukiji, diagnosis was confirmed with immunohistochemistry using CD56 Chuo-ku, Tokyo, 104-0045, Japan. Tel: +81 0335422511, Fax: +81 (1B6, 1:100; Novocastra, Newcastle upon Tyne, UK), chromogranin 0335453567, e-mail: [email protected] A (Polyclonal, 1:400; Dako, Glostrup, Denmark) and synaptophysin (Polyclonal, ready to use; Nichirei Biosciences, Tokyo, Japan) to Key Words: Cisplatin, esophageal cancer, extensive disease, exclude other esophageal subtypes of tumors. All patients had irinotecan hydrochloride, neuroendocrine carcinoma. received histological diagnoses from biopsy specimens; according

0250-7005/2014 $2.00+.40 5037 ANTICANCER RESEARCH 34: 5037-5042 (2014)

Table I. Patients’ characteristics and clinical outcomes.

No. Age, Stage Disease Metastatic IP treatment Second-line Later Response OS gender profile lesions cycles treatment treatment (months)

1 52, M IV ED HEP 3 EP wPTX, ADM+MMC PR 8.1 +5-FU, CPT-11 2 72, F III ED none 3 EP AMR, DTX PR 28.6 3 54, M IV ED HEP, LYM, OSS 1 – – PD 1.1 4 66, M rec ED HEP 5 DTX – SD 15.1 5 44, M IV ED PUL, HEP, LYM 9 CPT-11 EP PR 46.2 6 61, M IV ED HEP 2 DTX – PD 32.2 7 66, M rec ED LYM 1 – – NE 14.9 8 39, M rec ED HEP 1 CPT-11 AMR, wPTX SD 12.6 9 56, F IV ED HEP, stomach, LYM 1 – – NE 4.6 10 69, F IV ED LYM, MAR, crural muscle 3 AMR – PR 5.1 11 63, M IV ED HEP 4 AMR – PR 5.1 12 72, M IV ED HEP, LYM 2 – – PR 1.7*

M, Male; F, female; ED, extensive disease; rec, recurrent; HEP, liver; LYM, lymph node; OSS, bone; PUL, lung; MAR, bone marrow; EP, etoposide + cisplatin; DTX, ; CPT-11, irinotecan; AMR, amrubicin; wPTX, weekly ; ADM, adriamycin; MMC, ; PR, partial response; PD, progressive disease; SD, stable disease; NE, not evaluated; OS, overall survival. *Patient still alive at the time of writing.

to the 7th edition of the Union for International Cancer Control – criteria in accordance with the Common Terminology Criteria for American Joint Committee on Cancer (7th UICC-AJCC) (11) they Adverse Events version 4.0 (CTCAE v4.0) (13). were at Stage III or IV or had recurrent disease with metastasis at the time of diagnosis. Statistical methods. The Kaplan–Meier method was used to estimate Pre-treatment evaluation of all patients included the assessment PFS and OS. Statistical analysis was performed with JMP9 (SAS of laboratory data; staging was determined by computed Institute Inc., Cary, NC, USA). tomography (CT) of the chest and abdomen and by esophagoscopy with biopsy. We applied the tumor-node-metastasis (TNM) staging Results system to all cases according to the 7th UICC-AJCC classification. All patients treated with chemotherapy gave their informed consent. Patients’ clinical stages, histology, subsequent chemotherapy and Patients’ characteristics. A total of 12 patients (eight men, responses, as defined by radiographic imaging, were evaluated. Data four women) with ED-NEC of the esophagus were diagnosed on hematological and non-hematological toxicities associated with and treated with IP as first-line chemotherapy between chemotherapy were also collected. Medical records and laboratory January 2000 and April 2013. Their ages ranged from 39 to data on each patient were retrieved for analysis and assessment of 72 years, with a median age of 62 years. All patients had chemotherapy for NEC of the esophagus, and associated complications were also reviewed. clinical stage III or IV or recurrent disease at diagnosis (Table I). The stage III patient (No. 2) was defined as having Evaluation of efficacy and study end points. All patients were ED because of metastasis to the abdominal lymph nodes evaluated radiographically with CT scanning on a regular basis to outside the tolerable radiation field. On histology, poorly- confirm effectiveness or progression. Efficacy of treatment was differentiated NEC was confirmed by hematoxylin-eosin evaluated as follows: (i) response rate; (ii) PFS; (iii) OS. Response staining and immunohistochemical staining using CD56, rate was assessed with the Response Evaluation Criteria in Solid chromogranin A and synaptophysin. Tumors criteria version 1.1 (RECIST 1.1) (12). PFS was defined as the time from the first cycle of chemotherapy to the first clinical evidence of progressive disease or death from any cause. OS was Chemotherapy protocols. Patients were treated with defined as the time from the first cycle of chemotherapy to the time chemotherapy consisting of either cisplatin 80 mg/m2 on day of death from any cause or the last follow-up. Because of the 1 in combination with irinotecan 70 mg/m2 on days 1 and retrospective nature of the data, these relevant end points were 15, or cisplatin 60 mg/m2 on day 1 in combination with chosen to reflect clinical practice. irinotecan 60 mg/m2 on days 1, 8 and 15. Each cycle was Patients were assessed radiographically after at least two cycles repeated every 28 days (10, 14). of chemotherapy, but the duration of response was not confirmed.

Safety profile. Hematological and non-hematological adverse events Feasibilities of IP. A total of 34 cycles was administered related to chemotherapy were described by using common toxicity (median 2.5 cycles, range 1 to 9 per patient). Three patients

5038 Okuma et al: Irinotecan plus Cisplatin for Esophageal NEC

Figure 1. Kaplan-Meier progression-free survival curve showing a median of 4.0 months (95% confidence interval (CI), 0.9 to 7.6 months) (a), and overall-survival curve showing a median of 12.6 months (95% CI, 4.6 to 28.6 months) (b) for 12 extensive-disease esophageal neuroendocrine carcinoma patients treated with irinotecan plus cisplatin as first line chemotherapy.

(25%) had dose reductions, including a 20% dose reduction Table II. Toxicity of irinotecan-plus-cisplatin in treatment of of both cisplatin and irinotecan from the second cycle neuroendocrine carcinoma of the esophagus. onward in two patients owing to hematologic toxicity and Gr 1 Gr 2 Gr 3 Gr 4 Gr 3/4 (%) renal toxicity, respectively, and a 20% dose reduction of cisplatin in the ninth cycle in another patient due to renal Hematologic toxicities toxicity. Treatment was discontinued because of progressive Leukopenia 1 3 5 1 50 disease in seven patients (58%), adverse events in three Neutropenia 0 1 4 4 67 patients (25%), patient refusal in one (8%) and loss to Anemia 3 3 4 0 33 follow-up in one (8%). The three patients who discontinued Thrombocytopenia 3 0 2 2 33 chemotherapy because of adverse events consisted of one Non-hematologic toxicities patient with grade 3 diarrhea and two patients with renal Nausea 3 3 2 0 17 toxicity (grades 2 and 4). Eight of the 12 patients required Diarrhea 1 4 1 0 8.3 at least second-line chemotherapy or radiotherapy for Vomiting 3 3 0 0 0 Anorexia 4 4 3 0 25 local control. A median of 1 (0 to 3) subsequent Fatigue 4 3 1 0 8.3 chemotherapeutic regimen was given. Subsequent Alopecia 4 4 – – – chemotherapy included irinotecan monotherapy, EP, Febrile neutropenia – – 3 0 25 paclitaxel, amrubicin, and docetaxel, and adriamycin plus AST 6 1 1 0 8.3 mitomycin plus 5-FU (Table I). ALT 4 2 1 0 8.3 Creatinine 5 4 0 1 8.3 Clinical outcomes of IP treatment. Disease control was Neuropathy: sensory 0 0 0 0 0 observed in eight patients (67%): six had partial responses (50%) and two had stable disease (17%). Two patients had Gr, Grade; AST, aspartate transaminase; ALT, alanine transaminase. progressive disease. The median PFS was 4.0 months (95% CI, 0.9 to 7.6 months), whereas the median OS was 12.6 months (95% CI, 4.6 to 28.6 months) (Figure 1). After a median follow-up of 10.4 months, nine patients had died of (Table II). Grade 3/4 non-hematological toxicities included tumor progression, two were lost to follow-up and one anorexia (25% of patients), nausea (17%), diarrhea (8.3%) patient remained alive. and creatinine increase (8.3%). Three patients discontinued treatment because of toxicity (creatinine increase or Safety profile of IP. Among the grade 3/4 observed diarrhea), as mentioned previously. Febrile neutropenia was hematological toxicities, 50% of patients had leukopenia, seen in three patients (25%). No treatment-related deaths 67% neutropenia, 33% anemia and 33% thrombocytopenia were observed.

5039 ANTICANCER RESEARCH 34: 5037-5042 (2014)

Discussion (i.e. EP), most of the patients had limited disease, they include different gastrointestinal organs, the doses of IP IP as first-line therapy for ED esophageal NEC resulted in a varied widely and, finally, adverse events were not recorded. response rate of 50%, a median PFS of 4.0 months and a Therefore, in our analysis of 12 NEC patients, we restricted median OS of 12.6 months. IP appears to be a feasible the criterion of primary tumor to the esophagus and all chemotherapy for ED esophageal NEC with respect to patients received IP as first-line chemotherapy. The efficacy efficacy, toxicity and availability in the clinical setting. of chemotherapy in this study corresponded with those in Poorly-differentiated esophageal NEC accounts for other studies of esophageal NEC. However, because of approximately 0.05% to 7.6% of all esophageal cancers (4- differences in treatment scheduling, doses differed within our 8). Pathomorphologically, these tumors are divided into study according to the date of initiation of treatment. Other small- and large-cell carcinomas. Within these, there are limitations included the small patient number; moreover, Ki- tumors that are “pure” NECs and tumors in which 30% or 67 staining (MIB-1, 1:100; Dako, Glostrup, Denmark), more of the tumor has adenocarcinomatous features; the which is included in the WHO classification for NEC latter are referred to as mixed adenoneuroendocrine diagnosis, was performed in only one patient. carcinomas or MANECs. All 12 of our cases were pure Because IP is used mainly in Japan, this study offers a rare NECs. Since the histology of these tumors strongly analysis of the feasibility of this regimen alone in treating resembles that of small-cell cancers of the lung, the National ED esophageal NEC. However, considering the higher Comprehensive Cancer Network (NCCN) 2012 guidelines incidence of SCLC and the identical nature of the recommend treatment according to SCLC therapy. However, pathological features of ED esophageal NEC and SCLC, new due to the rarity of esophageal NEC, no phase III treatments for ED esophageal NEC will most likely -for the randomized trials have been conducted. The treatment time being- follow those for SCLC. Of note, our current data options remain at category 2A according to the NCCN were comparable to those from JCOG9511 of ED-SCLC guideline, implicating a uniform NCCN consensus that the suggesting that, although the locations of the primary tumors intervention is appropriately-based on lower-level evidence. differed, the NEC histological type responded similarly to Since the 1980s, Western countries have considered EP as the same chemotherapy used for SCLC. Furthermore, our standard treatment for ED-SCLC (15). The JCOG9511 study, data were comparable to – or even better than – historical however, resulted in significantly better OS in the IP arm than in treatment data for ED esophageal NEC, which have included the EP arm (12.8 vs. 9.4 months, p=0.002) in ED-SCLC (10). chemotherapy regimens other than IP or EP. Therefore, at Therefore, in light of the fact that IP is currently the standard present, IP or EP is the recommended chemotherapy for ED first-line chemotherapy for ED-SCLC in Japan, we treated esophageal NEC. For better future treatment options, we patients with IP as first-line chemotherapy for ED esophageal believe it is important to know the exact efficacy and toxicity NEC. Our results should be of benefit to future investigations of the IP regimen when cases are restricted to ED esophageal of the use of SCLC regimens to treat NEC in other organs. NEC. Further research may need to focus on whether IP or In a retrospective Japanese multicenter analysis of IP or EP is the true beneficial first-line regimen for GI-NEC. EP chemotherapy for GI-NEC, out of 258 patients enrolled, 85 (33%) had primary lesions in the esophagus. Of all the References patients who received chemotherapy, 52% received IP. The 1 Bosman FT, Carneiro F, Hruban RH and Theise ND: World response rates of GI-NEC patients treated with IP was 51% Health Organization Classification of Tumours of the Digestive and the median OS was 13.4 months. The median OS for System, 4th Edition. IARC Press, Lyons France, 2010. esophageal NEC has been recorded at 13.4 months (16). In a 2 Ito T, Sasano H, Tanaka M, Osamura RY, Sasaki I, Kimura W, retrospective literature analysis of 199 esophageal small-cell Takano K, Obara T, Ishibashi M, Nakao K, Doi R, Shimatsu A, carcinoma patients, including 93 (46.7%) with limited Nishida T, Komoto I, Hirata Y, Nakamura K, Igarashi H, Jensen disease and 95 (47.7%) with ED, the median OS was 8 RT, Wiedenmann B and Imamura M: Epidemiological study of gastroenteropancreatic neuroendocrine tumors in Japan. J months for the former and 3 months for the latter (17). In Gastroenterol 45: 234-243, 2010. another single-Institution study of 12 esophageal small-cell 3 Scoazec JY, Couvelard A, Monges G, Leteurtre E, Belleannee carcinoma patients (5 with limited disease and 7 with ED) G, Guyetant S, Duvillard P, Danjoux M, Parot X and Lepage C: treated with IP, the response rate was 83% with a median OS Well-differentiated grade 3 digestive neuroendocrine tumors: of 13.9 months (18). Other single-Institution studies have Myth or reality? The PRONET study group. J Clin Oncol 30 reported on 12 and 8 cases of esophageal small cell (suppl; abstr 4129), 2012. carcinoma with a median OS of 7 and 12.5 months, 4 Maru DM, Khurana H, Rashid A, Correa AM, Anandasabapathy S, Krishnan S, Komaki R, Ajani JA, Swisher SG and Hofstetter respectively (6, 19). However, all of the historical WL: Retrospective study of clinicopathologic features and retrospective studies have limitations in that some were prognosis of high-grade neuroendocrine carcinoma of the conducted in multiple institutions with different regimens esophagus. Am J Surg Pathol 32: 1404-1411, 2008.

5040 Okuma et al: Irinotecan plus Cisplatin for Esophageal NEC

5 Law SY, Fok M, Lam KY, Loke SL, Ma LT and Wong J: Small 15 Roth BJ, Johnson DH, Einhorn LH, Schacter LP, Cherng NC, cell carcinoma of the esophagus. Cancer 73: 2894-2899, 1994. Cohen HJ, Crawford J, Randolph JA, Goodlow JL, Broun GO et 6 Huncharek M and Muscat J: Small cell carcinoma of the al: Randomized study of , , and esophagus: The Massachusetts General Hospital experience, versus etoposide and cisplatin versus alternation of 1978 to 1993. Chest 107: 179-181, 1995. these two regimens in extensive small-cell lung cancer: a phase 7 Doherty MA, McIntyre M and Arnott SJ: Oat cell carcinoma of III trial of the Southeastern Cancer Study Group. J Clin Oncol the esophagus: a report of six British patients with a review of 20: 282-291, 1992. the literature. Int J Radiat Oncol Biol Phys 10: 147-152, 1984. 16 Machida N, Yamaguchi T, Kasuga A, Takahashi H, Sudo K, 8 Briggs MF and Ibrahim NB: Oat cell carcinomas of the oesophagus: Nishina T, Tobimatsu K, Ishido K, Furuse J and Boku N: a clinico-pathological study of 23 cases. Histopathology 7: 261-277, Multicenter retrospective analysis of systemic chemotherapy for 1983. advanced poorly differentiated neuroendocrine carcinoma of the 9 Travis WD, Brambilla E, Muller-Hermelink HK and Harris CC: digestive system. J Clin Oncol 30 (suppl; abstr 4046), 2012. World Health Organization Classification of Tumours. Pathology 17 Casas F, Ferrer F, Farrús B, Casals J and Biete A: Primary small and Genetics of Tumours of the Lung, Pleura, Thymus and cell carcinoma of the esophagus: a review of the literature with Heart. IARC Press, Lyon France, 2004. emphasis on therapy and prognosis. Cancer 80(8): 1366-1372, 1997. 10 Noda K, Nishiwaki Y, Kawahara M, Negoro S, Sugiura T, 18 Chin K, Baba S, Hosaka H, Ishiyama A, Mizunuma N, Yokoyama A, Fukuoka M, Mori K, Watanabe K, Tamura T, Shinozaki E, Suenaga M, Kozuka T, Seto Y, Yamamoto N and Yamamoto S and Saijo N: Irinotecan plus cisplatin compared Hatake K: Irinotecan plus cisplatin for therapy of small-cell with etoposide plus cisplatin for extensive small-cell lung cancer. carcinoma of the esophagus: report of 12 cases from single N Engl J Med 346: 85-91, 2002. institution experience. Jpn J Clin Oncol 38: 426-431, 2008. 11 Sobin LH, Gospodarowicz MK, Wittekind C and International 19 Medgyesy DC, Wolff RA, Putnam Jr. JB and Ajani JA. Small Union against Cancer: TNM classification of malignant tumors. cell carcinoma of the Esophagus. The University of Texas MD 7. Wiley-Blackwell: Oxford, 2010. Anderson Cancer Center experience and literature review. 12 Eisenhauer EA, Therasse P, Bogaerts J, Schwartz LH, Sargent Cancer 88: 262-267, 2000. D, Ford R, Dancey J, Arbuck S, Gwyther S, Mooney M, Rubinstein L, Shankar L, Dodd L, Kaplan R, Lacombe D and Verweij J: New response evaluation criteria in solid tumors: revised RECIST guideline (version 1.1). Eur J Cancer 45: 228- 247, 2009. 13 National Cancer Institute: Common Terminology Criteria for Adverse Events v.4.0 (CTCAE): http://ctep.cancer.gov/protocol Development/electronic_applications/ctc.htm#ctc_402011. 14 Boku N, Yamamoto S, Fukuda H, Shirao K, Doi T, Sawaki A, Koizumi W, Saito H, Yamaguchi K, Takiuchi H, Nasu J and Ohtsu A: versus combination of irinotecan plus Received May 9, 2014 cisplatin versus S-1 in metastatic gastric cancer: a randomized Revised June 28, 2014 phase 3 study. Lancet Oncol 10: 1063-1069, 2009. Accepted July 1, 2014

5041