ORIGINAL ARTICLE

Definitive Chemoradiation With Full-dose Gemcitabine for Unresectable Pancreatic Cancer Efficacy of Involved-Field Radiotherapy Jessica Zhou, MD,* Charles X. Fang, BS,w Jincheng Shen, MS,z Matthew J. Schipper, PhD,z Mark M. Zalupski, MD,y Rebecca M. Minter, MD,8 Diane M. Simeone, MD,8 Deanna Hart, BS,w Taylor Eagle, BS,w Grace W. Kimmel, BS,w Jeffrey Vainshtein, MD,* Theodore S. Lawrence, MD, PhD,* and Mary Feng, MD*

Objectives: Definitive chemoradiotherapy for unresectable pancreatic cancer has traditionally involved 5-fluorouracil-based chemotherapy. Our institution has a long history of combining gemcitabine and radiotherapy (RT), and performed a retrospective review of all patients treated in this manner. Materials and Methods: We reviewed the records of 180 patients treated from 1999 to 2012. Mean RT dose was 40.9 Gy in 2.2-Gy fractions, and targeted only radiographically apparent disease. Ninetysix percent of patients received full-dose gemcitabine-based chemotherapy with RT. Kaplan-Meier was used to analyze time-to-event endpoints, and Cox regression models were used to assess significant prognostic variables. Results: Eighty-nine percent of patients completed RT without a toxicity-related treatment break. Median follow-up was 10.2 months. Twenty-nine percent of patients had a radiographic decrease in primary tumor size following treatment. Median overall survival was 11.8 months, time to distant metastasis (TDM) was 6.7 months, and time to local recurrence (TLR) was 8.3 months. On multivariate analysis, male sex, lower performance status, and higher posttreatment CA 19-9 level predicted for worse overall survival. Posttreatment, CA 19-9 was also associated with TDM and TLR, and radiographic tumor response was associated with better TLR. Conclusion: Definitive chemoradiation using full-dose gemcitabine is well tolerated and achieves survival outcomes comparable to reported trials in the literature. Key Words: pancreatic cancer, unresectable, radiation therapy, gemcitabine

(Am J Clin Oncol 2015;00:000–000)

P

ancreatic cancer is the fourth leading cause of cancer death, with an estimated 40,560 deaths and 48,960 new diagnoses in the United States in 2015, and its incidence has increased

From the Departments of *Radiation Oncology; zBiostatistics; 8Surgery, University of Michigan; yDepartment of Internal Medicine, Division of Hematology/Oncology, University of Michigan; and wUniversity of Michigan Medical School, Ann Arbor, MI. J.Z. and C.X.F. contributed equally to this work as co-first authors. The authors declare no conflicts of interest. Reprints: Mary Feng, MD, Department of Radiation Oncology, University of Michigan, 1500 E. Medical Center Dr., UH B2C 490, Ann Arbor, MI 48109. E-mail: [email protected]. Copyright r 2015 Wolters Kluwer Health, Inc. All rights reserved. ISSN: 0277-3732/15/000-000 DOI: 10.1097/COC.0000000000000200

over the past 10 years.1 The prognosis for advanced unresectable disease remains quite poor, with a median survival of 8 months to 1 year. Treatment options include radiation, chemotherapy, or concurrent chemoradiation, which has been shown to be superior to chemotherapy alone.2 Historically, radiation therapy has been directed to the primary tumor and all regional lymph nodes, combined with 5-fluorouracil (FU) chemotherapy. More recently, gemcitabine has been increasingly used as the chemotherapeutic agent after it was shown to improve symptomatic relief and survival for advanced symptomatic patients.3 Because of gemcitabine’s potent radiosensitizing effects, initial studies have shown significant toxicity with high-dose therapy combined with extended-field radiotherapy (RT).4,5 Treatment with a decreased dose of 600 mg/m2 and concurrent RT has shown to be safe and improved survival compared with gemcitabine alone in a recent randomized trial.6 Nevertheless, tumor response rates continue to be low, and both local and systemic failures are significant.7 Treatment intensification with full-dose concurrent gemcitabine has become an area of ongoing research to exploit gemcitabine’s full systemic effect. To limit toxicity, radiation treatment fields have been reduced to cover only gross radiographic disease without prophylactic nodal irradiation, relying on gemcitabine to eliminate microscopic disease. As shown in a phase II clinical trial, this approach has yielded acceptable overall survival (OS) and treatment tolerance, with a very low incidence of failure in regional nodes not covered in the RT field.8,9 Clinical experience using full-dose gemcitabine in combination with definitive RT is still limited, and only a few institutions have reported their experience thus far.10,11 Our institution previously reported a series of 74 patients who were treated with this regimen,9 which showed limited toxicity and a median OS rate of 11.2 months. In this study, we have consolidated all locally advanced unresectable patients who were treated either on or off clinical protocols throughout the years. We aim to evaluate treatment efficacy and identify predictors of clinical outcome using full-dose gemcitabine chemoradiotherapy (chemoRT).

MATERIALS AND METHODS Patient Selection After institutional review board approval, the records of patients who underwent definitive treatment with RT at the University of Michigan from February 1999 to June 2012 were

1 Copyright r 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited. American Journal of Clinical Oncology



Volume 00, Number 00, ’’ 2015

www.amjclinicaloncology.com |

Zhou et al

American Journal of Clinical Oncology

reviewed. All patients were judged to be unresectable or borderline resectable at the time of diagnosis by the surgeon or multidisciplinary tumor board based on radiologic findings of tumor involvement of the celiac axis, superior mesenteric artery, or extensive involvement of the superior mesenteric vein not amenable to vascular bypass. All borderline patients who subsequently underwent curative surgery after CRT were excluded (n = 18). All patients who had radiation treatment at an outside facility, did not complete at least 80% of their prescribed course of radiation treatment, did not receive concurrent chemotherapy with radiation, received concurrent non–gemcitabine-based chemotherapy, or had evidence of metastasis on imaging before starting treatment were excluded. The final cohort for analysis included 180 patients.

Treatment As part of pretreatment evaluation, all patients underwent complete history and physical examination, routine laboratory studies, helical computed tomography (CT) scan of the abdomen with intravenous contrast, chest CT, endoscopic ultrasound or endoscopic retrograde cholangiopancreatography, and biopsy. All patients underwent biopsy of the tumor before initiation of treatment and had a definitive diagnosis of adenocarcinoma. All pathology obtained at outside institutions was reviewed at our institution. Unresectability or borderline resectability was determined by CT, endoscopic ultrasound, endoscopic retrograde cholangiopancreatography, or abdominal magnetic resonance imaging. Patients who had evidence of jaundice or elevated bilirubin levels underwent placement of a biliary stent before assessment of CA19-9 levels and initiation of treatment.



Volume 00, Number 00, ’’ 2015

Chemotherapy As part of the chemotherapy regimen given concurrently with RT, 120 patients (67%) received gemcitabine alone, 35 (19%) received gemcitabine and oxaliplatin, 17 (9%) received gemcitabine and cisplatin, and 8 (4%) received gemcitabine and oral 5-FU (Xeloda). For the majority of patients (96%), the gemcitabine was given full dose for 1 or 2 cycles (1000 mg/m2 over 30 min intravenously on days 1 and 8 of a 21-d cycle) depending on the length of RT. Of the rest who did not receive full-dose treatment concurrently with RT, the minimum gemcitabine dose received was 500 mg/m2. In addition, 90 patients (50%) received at least 1 cycle of chemotherapy before start of RT, with 30 patients (17%) receiving >1 cycle (median, 1 cycle; range, 1 to 7 cycles). The majority of those who underwent neoadjuvant chemotherapy (92%) were treated with a gemcitabine-based regimen, with the rest receiving FOLFIRINOX (5-FU with leucovorin, oxaliplatin, irinotecan) or FOLFOX (5-FU with leukovorin, oxaliplatin), 84% of all patients received additional adjuvant chemotherapy post-RT, for a median of 2 cycles (range, 1 to 12 cycles). All post-RT chemotherapy was gemcitabine based except in 1 case where FOLFIRINOX was given. Patients were scheduled posttreatment follow-up visits every 3 to 4 months, and CT imaging was used to evaluate for local progression or metastasis. Toxicity information was collected prospectively for patients who were enrolled on clinical protocols, and retrospectively at routine follow-up visits for patients who were not enrolled on protocols. Toxicity was scored according to the National Cancer Institute’s Common Terminology Criteria for Adverse Events.

RT

Statistical Analysis

Radiation therapy was planned with 6 to 16 MV x-rays, on contrast-enhanced CT simulation scans. A total of 115 patients (64%) were treated with 3-dimensional conformal RT, and 65 (36%) were treated with intensity-modulated RT (IMRT). The gross target volume was limited to only radiographically apparent disease including the primary tumor and malignant-appearing regional lymph nodes without elective nodal coverage with malignancy as determined from radiology reports, including nodes of abnormal size, morphology, and/or enhancement. 5 mm were added for microscopic disease clinical target volume. PTV margin has evolved over the years with changes in tumor immobilization and image guided radiation therapy, from 7 to 10 mm initially before Active Breathing Control and cone beam CT to our current practice of 5 mm with these motion management techniques, with an additional customized margin for motion obtained from 4demensional CT for patients unable to cooperate with Active Breathing Control. Target dose ranged from 24 to 60 Gy in 1.8 to 2.8 Gy per fraction, with median and mean total delivered dose of 36 and 40.9 Gy in 2.2 Gy per fraction. Mean total bioequivalent dose in 2-Gy fractions (EQD2) was 49.9 Gy (median, 44.6 Gy; range, 27.8 to 74.4 Gy). Patients were enrolled on multiple prospective clinical protocols throughout the years. In the early era (1999 to 2006), patients were treated predominantly with 15 fractions ranging from 1.8 to 2.8 Gy per fraction (most commonly, 2.0 to 2.4). In the later era (2007 to 2012), patients were treated predominantly with 25 fractions ranging from 2.0 to 2.4 Gy per fraction (most commonly, 2.1 to 2.3). Treatment plans for all patients were generated in our inhouse planning system, UM Plan. Analysis of the dose-volume histogram data was performed in Igor Pro (version 6.2.1.0; Wave Metrics Inc., Lake Oswego, OR).

OS, time to local progression, and time to distant metastases were calculated from the start of RT. The Kaplan-Meier method was used to summarize time-to-event endpoints, and the log-rank test was used to compare survival curves between groups. Cox regression models were used to assess the effect of variables on survival. Logistic regression was used to assess the effect of variables on the development of metastases. CA 19-9 values were log-transformed before inclusion in regression models because of an approximately logarithmic relationship between OS and CA19-9. Patients with CA 19-9 levels of Normal to

Definitive Chemoradiation With Full-dose Gemcitabine for Unresectable Pancreatic Cancer: Efficacy of Involved-Field Radiotherapy.

Definitive chemoradiotherapy for unresectable pancreatic cancer has traditionally involved 5-fluorouracil-based chemotherapy. Our institution has a lo...
141KB Sizes 4 Downloads 8 Views