Paclitaxel and concurrent radiation for locally advanced pancreatic cancer

Int. J. Radiation Oncology Biol. Phys., Vol. 49, No. 5, pp. 1275–1279, 2001 Copyright © 2001 Elsevier Science Inc. Printed in the USA. All rights reserved 0360-3016/01/$–see front matter

PII S0360-3016(00)01527-3

CLINICAL INVESTIGATION

Pancreas

PACLITAXEL AND CONCURRENT RADIATION FOR LOCALLY ADVANCED PANCREATIC CANCER HOWARD SAFRAN, M.D.,* TODD MOORE, M.D.,* DAVID IANNITTI, M.D.,* TOM DIPETRILLO, M.D.,* PAUL AKERMAN, M.D.,* WILLIAM CIOFFI, M.D.,* DAVID HARRINGTON, M.D.,* DANIEL QUIRK, M.D.,* RATESH RATHORE, M.D.,* DENNIS CRUFF, M.D.,* JAMSHEED VAKHARIA, M.D.,* SUJAY VORA, M.D.,* DIANNE SAVARESE, M.D.,† AND HAROLD WANEBO, M.D.* *The Brown University Oncology Group, Providence, RI; †The University of Massachusetts Medical Center, Worcester, MA Purpose: To determine the activity and toxicity of paclitaxel and concurrent radiation for pancreatic cancer. Methods and Materials: Forty-four patients with locally unresectable pancreatic cancer were studied. Patients received paclitaxel, 50 mg/m2 by 3 h i.v. (IV) infusion, weekly, on Days 1, 8, 15, 22 and 29. Radiation was administered concurrently to a total dose of 50.4 Gy, in 1.80 Gy fractions, for 28 treatments. Results: Nausea and vomiting were the most common toxicities, Grade 3 in five patients (12%). Two patients (5%) had Grade 4 hypersensitivity reactions to their first dose of paclitaxel. Of 42 evaluable patients, the overall response rate was 26%. The median survival was 8 months, and the 1-year survival was 30%. Conclusion: Concurrent paclitaxel and radiation demonstrate local-regional activity in pancreatic cancer. Future investigations combining paclitaxel with other local-regional and systemic treatments are warranted. © 2001 Elsevier Science Inc. Paclitaxel, Radiotherapy, Pancreatic cancer.

INTRODUCTION

cells that are resistant to paclitaxel as a single agent (9). Furthermore, stem cells of the gastrointestinal mucosa are not substantially radiosensitized by paclitaxel (10). This may improve the therapeutic index of paclitaxel/RT when the upper abdomen is included within the radiation field. The Brown University Oncology Group completed a Phase I study of paclitaxel/RT for locally advanced pancreatic and gastric cancer (4). The maximum tolerated dose of paclitaxel was 50 mg/m2/week for 6 weeks with abdominal radiation. The dose-limiting toxicities were abdominal pain within the RT field, nausea, and anorexia and were attributable to paclitaxel-induced RT enhancement. Promising activity was noted in this Phase I study. We therefore performed a Phase II study to determine the response rate and toxicity of paclitaxel/RT for pancreatic cancer.

There were approximately 29,000 deaths from pancreatic carcinoma in 1998 in the United States. The overall 5-year survival has remained constant at ⬍5% (1). Approximately half of all patients have locally advanced disease at the time of initial diagnosis. Standard treatment for locally unresectable disease is fluorouracil (5-FU) and external beam irradiation (2). The addition of 5-FU as a radiosensitizing agent modestly increases local control and median survival; however, virtually all patients will eventually develop disease progression and death (3). We sought to develop an active local-regional therapy for pancreatic cancer while still recognizing the need for improvements in preventing systemic relapse. The Brown University Oncology Group (BrUOG) has evaluated paclitaxel as a radiation sensitizer to attempt to improve local-regional control in pancreatic cancer (4). Paclitaxel synchronizes cells at G2/M, a relatively radiosensitive phase of the cell cycle (5–7). In addition, response to paclitaxel and concurrent radiation (paclitaxel/RT) is not affected by p53 mutations, suggesting that paclitaxel/RT is a rational treatment approach for malignancies that frequently harbor p53 mutations, such as pancreatic cancer (8). Paclitaxel can enhance the cytotoxicity of RT even in cancer

METHODS AND MATERIALS Eligibility Patients with histologically documented adenocarcinoma of the pancreas were enrolled in this study from March 1995 to October 1999. Eligible patients included those with locally unresectable disease due to vascular invasion, regional adenopathy, or medical contraindications to resection. Patients were required to have measurable or assessable dis-

Reprint requests to: Howard Safran, M.D., Department of Medicine, The Miriam Hospital, 164 Summit Ave., Providence, RI 02906.

Supported in part by Bristol-Myers Squibb Company. Accepted for publication 19 October 2000. 1275

1276

I. J. Radiation Oncology

●

Biology

●

Physics

ease. Patients with known metastatic disease to other organs, ascites, or peritoneal implants and those who had received prior paclitaxel or irradiation to the planned field were ineligible. Eligibility criteria included age ⱖ18. Required laboratory parameters included granulocyte count ⱖ1,800 cells/␮L, platelet count ⱖ100,000 cells/␮L, bilirubin ⱖ3 mg/dL, and creatinine ⱖ3 mg/dL. Patients with biliary or gastroduodenal obstruction must have had prior drainage before starting chemoradiation. A complete history and physical examination were performed on all patients before treatment. Height, weight, performance status, and tumor stage were recorded. Required staging studies included a chest radiograph and abdominal computed tomographic (CT) scan. Laparoscopy was not required. All patients gave written informed consent according to federal and institutional guidelines. Chemotherapy Paclitaxel, 50 mg/m2/week, was administered as a 3-h i.v. (IV) infusion for 6 weeks on Days 1, 8, 15, 22, 29, and 36. Paclitaxel was generally given at the beginning of the week before radiation treatment. Dexamethasone 20 mg IV, diphenhydramine 25 mg IV, and ranitidine 50 mg IV were given 30 min before paclitaxel. Dosages of dexamethasone and diphenhydramine could be reduced in subsequent weeks if no hypersensitivity reactions were observed. The initial protocol mandated that patients receive no further therapy after paclitaxel/RT until disease progression. However, when the matrix metalloproteinase inhibitor marimastat became available on clinical trial to the BrUOG, 12 patients subsequently received marimastat, 10 mg BID, after paclitaxel/RT in a sequential Phase II trial. Radiation therapy Radiation therapy was delivered to 50.4 Gy in 28 fractions of 1.8 Gy per fraction to the tumor and draining lymph nodes over 5 weeks. The initial volume radiated to 45 Gy was defined by the top of the T11 vertebral body and the bottom of L3. All other margins were defined by a 2-cm margin on all gross disease. An additional reduced field treated gross disease with a 1–1.5-cm margin to 5.4 Gy. Multiple field techniques were used with 10 –25 mv photons, though four-field technique was common. CT scans in the treatment position were used to identify appropriate anatomy. No more than 30% of the total kidney volume received ⬎50% of the prescribed dose. Spinal cord dose was maintained below 45 Gy. No patient exceeded these planning guidelines. Dose modifications for toxicity Paclitaxel dosage was reduced by 50% for ANC 1,000 – 1,799/␮L or platelet count 75,000 –99,000/␮L. Paclitaxel was omitted for ANC ⬍ 1,000/␮L or platelet count ⬍75,000/␮L until hematologic recovery. Radiation was held for ANC ⬍ 500/␮L or platelet count ⬍ 50,000/␮L. Paclitaxel and radiation were held for Grade 3 or Grade 4

Volume 49, Number 5, 2001

Table 1. Patient characteristics (n ⫽ 44) Age Median Range Sex (M:F) Performance status 0 1 2 Clinical stage T2N0 T4N0 T1–4N1 Local recurrence Tumor size ⱖ5 cm ⬍5 cm

64 42–85 26:19 7 30 7 1 27 14 2 13 31

nonhematologic toxicity and not resumed until recovery to Grade 2 or better. Paclitaxel dosage was then reduced by 50%. Toxicity and response criteria A CT scan was repeated 4 – 8 weeks after completion of paclitaxel/RT. A complete response required no evidence of tumor on post-treatment CT scan. A partial response required a 50% reduction in the product of the maximum perpendicular tumor measurements. For patients with assessable but nonmeasurable disease by CT scan, a partial response was defined as a 50% decrease in tumor mass as assessed by CT scan. Toxicity was graded by Cancer and Leukemia Group B (CALGB) toxicity criteria. Statistical analysis Survival curves were calculated according to the method of Kaplan-Meier. The registration date was taken as the starting point of patient time in the analysis. RESULTS Patient characteristics Forty-five patients were entered on this study; however, one patient was ineligible due to metastatic disease. Characteristics of the 44 eligible patients are listed in Table 1. The mean age was 64, and three patients were older than 80 years. Forty-one patients were unresectable due to vascular invasion or regional adenopathy, two patients had local recurrences following previous surgery, and one had potentially resectable disease by CT scan but had a medical contraindication to surgery. Eleven patients had either a staging laparoscopy or exploratory laparotomy before chemoradiation. Thirty-nine patients had tumors of the head, 4 had tumors confined to the body, and 1 in the tail of the pancreas. Tumor size was ⱖ5 cm in 13 patients; 31 patients had tumors ⬍5 cm. Twenty-four patients had endobiliary stents, 9 had enterobiliary bypass surgery, 2 had percutaneous biliary drains, and 9 patients had no obstructive jaundice and were not bypassed before chemoradiation.

Paclitaxel/RT for pancreatic cancer

● H. SAFRAN et al.

1277

Table 3. Response (n ⫽ 42) Partial response Stable disease Progression

11 (26%) 21 (50%) 10 (24%)

toxicities were most common in the fifth and sixth weeks of treatment. Two patients (5%) had Grade 3 epigastric pain that was presumed to be from chemoradiation-induced gastritis. Two patients (5%) had Grade 4 hypersensitivity reactions to their first paclitaxel treatment and were not retreated. There were no toxic deaths. No significant renal or hepatic toxicity was observed. Excluding the two patients with hypersensitivity reactions, 99% percent of the planned radiation and 92% of planned chemotherapy were administered.

Fig. 1. Typical AP treatment field for a lesion of the pancreatic head.

Anterior field and posterior field width ranged from 12– 16.5 cm and length from 13–17 cm. Approximately 25% of the area was blocked. Lateral field width ranged from 9 –14 cm, with an average of 20% of the field being blocked. A typical radiation therapy field is demonstrated in Fig. 1. Toxicity of chemoradiation Acute toxicities during chemoradiation are listed in Table 2. Multiple toxicities in the same patient are scored as separate events. Ten of 44 patients (23%) had Grade 3 or greater toxicity. Nausea and vomiting were the most common toxicity; it was Grade 3 in five patients (12%). There were three instances of Grade 4 neutropenia (7%), and one patient (2%) had Grade 3 thrombocytopenia. Hematologic

Response Excluding the two patients removed from this study due to hypersensitivity reactions, 42 patients were evaluable for response and survival. Response rates are listed in Table 3. Eleven of 42 patients (26%) had a partial response. Fourteen patients who were initially unresectable, with stable or responding disease by CT, underwent surgical exploration: 4 were resected, 1 had fibrosis with negative biopsy, and 9 had unsuspected liver metastases. Of the patients who were resected, 3 had partial responses, and 1 had stable disease following chemoradiation. Survival Thus far, 40 patients have died, all from metastatic cancer. Two patients remain alive without evidence of disease at 53 and 7 months, respectively. Of 42 patients that progressed, only 2 had an isolated local recurrence as their first site of progression. Two additional patients had local and systemic failure. Figure 2 shows the overall survival. The median survival was 8 months for all patients and 17 months for partial responders. The one-year survival was 30%. DISCUSSION

Table 2. Toxicity (n ⫽ 44) Grade

Nausea/vomiting Pain/gastritis Neutropenia Anemia Hypersensitivity Dehydration Thrombocytopenia Infection Anorexia Diarrhea Hypotension

2

3

4

10 (24%) 6 (14%) 8 (19%) 6 (14%) – 2 (5%) 1 (2%) 3 (7%) 2 (5%) – –

5 (12%) 2 (5%) – 1 (2%) – 1 (2%) 1 (2%) 1 (2%) 1 (2%) 1 (2%) 1 (2%)

– – 3 (7%) – 2 (5%) – – – – – –

Locally advanced pancreatic cancer is characterized by local and systemic failure, and conventional therapies are generally ineffective in preventing disease progression. We have studied paclitaxel as a radiation sensitizer in pancreatic cancer. Our goal was to use paclitaxel/RT as a building block upon which future radiation sensitizers could be added to improve local control; novel systemic agents could be incorporated to delay systemic progression. A Phase II study of single-agent paclitaxel in pancreatic cancer demonstrated an 8% response rate, including one complete response, in 45 patients (11). While this is comparable to the single-agent response rate of gemcitabine (12–14), we chose to exploit paclitaxel’s ability to function as a radiosensitizer. Nausea, vomiting, and gastritis were the

1278

I. J. Radiation Oncology

●

Biology

●

Physics

Volume 49, Number 5, 2001

Fig. 2. Overall survival with 95% confidence interval.

most important toxicities in this study. These toxicities appeared in part due to inflammation and edema from paclitaxel-induced radiation enhancement. Patients’ performance status appeared to correlate with toxicity. The patients accrued to this study had very extensive pancreatic cancers. Often, these patients had pretreatment anorexia and nausea and thus were prone to toxicity. Despite this, none of the patients required total parenteral nutrition, and prophylactic enteral feeding tubes were not used. Paclitaxel/RT therefore was a relatively simple outpatient treatment. Substantial local-regional activity was demonstrated in this study. The overall response rate was 26%, with a median survival of 17 months for responding patients. Response to therapy could not be predicted by tumor size, tumor location, presence of nodal involvement, or treatment intensity. Since systemic failure was dominant, accurate determination of local progression was difficult to ascertain. Maximal radiographic response was usually demonstrated by CT scan 8 weeks following completion of chemoradiation. Since radiographic responses can be difficult to assess in locally advanced pancreatic cancer following chemoradiation, it may have been informative to monitor CA19 –9 before and following completion of chemoradiation. CA19 –9 monitoring was not routine when this study began in 1995, and therefore it was not included in this study.

However, CA19 –9 monitoring has been incorporated into future BrUOG trials to determine if CA19 –9 predicts response, time to progression, and survival. Since the focus of our trial was on local response and toxicity, we allowed the inclusion of patients with large tumors ⬎5 cm and extensive regional and portal adenopathy. These patients were prone to rapid systemic relapse, which contributed to the relatively short median survival. If laparoscopy had been required, it may have been possible to identify and exclude patients with peritoneal or hepatic implants that may have improved survival rates. This is the first report evaluating paclitaxel as a radiosensitizer in pancreatic cancer. The main focus of our trial was on local activity and toxicity; survival was heavily dependent on systemic relapse. The Radiation Therapy Oncology Group (RTOG) has recently completed a Phase II study of paclitaxel/RT for locally advanced pancreatic cancer. Response and survival data should be available within the next one to two years. It is difficult to compare survival data between our study of paclitaxel/RT and historical Phase II studies using 5-FU as a radiation sensitizer, however, survival appears comparable. The Gastrointestinal Study Group (GITSG) demonstrated an increase in survival from 5.9 months to 13.0 months with radiation alone as compared to radiation and 5-FU (3). Importantly, the patients in the GITSG had ab-

Paclitaxel/RT for pancreatic cancer

dominal surgery establishing regional localization. Comparable median and one-year survivals have also been reported with 5-FU and radiation by Mohiuddin et al. and Whittington et al. (15, 16). A Phase II study of cisplatin, 5-FU and radiation followed by additional post chemotherapy 5-FU and leucovorin has been reported by Andre et al. showing substantially more toxicity and a similar 9-month median and 31% one-year survival (17). Gemcitabine has recently shown promise as a radiation sensitizer (18 –20); however, since Phase II data in pancreatic cancer are not yet available, it is not possible to compare paclitaxel and gemcitabine as radiation sensitizers. To improve loco-regional activity, we have initiated a Phase I/II study of gemcitabine, paclitaxel and radiation for locally advanced pancreatic cancer. In our trial, of the 42 recurrences, only two patients had isolated local recurrences as the first site of progression, and

● H. SAFRAN et al.

1279

two had local and systemic recurrences. Therefore, to ultimately improve survival, more active systemic therapy to prevent the growth of systemic metastases will have to be integrated before or following chemoradiation. This has prompted us to begin evaluating novel noncytotoxic agents, such as marimastat, following completion of chemoradiation (21). The median survival of patients treated with marimastat was 13 months, however only patients who responded or had stable disease after chemoradiation were eligible to receive marimastat. A prospective randomized treatment trial would be needed to determine whether marimastat could delay relapse and improve survival. Farnesyl transferase inhibitors, angiogenesis inhibitors and monoclonal antibodies directed against growth factor receptors are among other potential agents that may be investigated following effective chemoradiation in locally advanced pancreatic cancer (21).

REFERENCES 1. Landis SH, Murray T, Bolden S, et al. Cancer statistics, 1998. CA Cancer J Clin 1998;48:6 –29. 2. Lillemoe KD. Current management of pancreatic carcinoma. Ann Surg 1995;221:133–148. 3. Moertel CG, Frytak S, Hahn RG, et al. Therapy of locally unresectable pancreatic carcinoma: A randomized comparison of high dose (6000 rads) radiation alone, moderate dose radiation (4000 rads ⫹ 5-fluorouracil), and high dose radiation ⫹ 5-fluorouracil: The Gastrointestinal Tumor Study Group. Cancer 1981;48:1705–1710. 4. Safran H, King T, Choy H, et al. Paclitaxel and concurrent radiation for locally advanced pancreatic and gastric cancer: A Phase I study. J Clin Oncol 1997;15:901–907. 5. Geard CR, Jones JM, Schiff PB. Taxol and radiation. J Natl Cancer Inst 1993;5:89 –94. 6. Tishler R, Schiff P, Geard C, et al. Taxol: A novel radiation sensitizer. Int J Radiat Oncol Biol Phys 1992;22:613– 617. 7. Choy H, Rodriguez FF, Koester S, et al. Investigation of Taxol as a potential radiation sensitizer. Cancer 1993;71: 3774 –3778. 8. Safran H, King T, Choy H, et al. p53 mutations do not predict response to paclitaxel/radiation for non-small cell lung cancer. Cancer 1996;78:1209 –1216. 9. Saito Y, Milross CG, Hittelman WN, et al. Effect of radiation and paclitaxel on p53 expression in murine tumors sensitive or resistant to apoptosis induction. Int J Radiat Oncol Biol Phys 1997;38:623– 631. 10. Mason KA, Milas L, Peters LJ. Effect of paclitaxel (Taxol) alone and in combination with radiation on the gastrointestinal mucosa. Int J Radiat Oncol Biol Phys 1997;38:623– 631. 11. Whitehead RP, Jacobson J, Brown TD, et al. Phase II trial of paclitaxel and granulocyte colony-stimulating factor in patients with pancreatic carcinomas: A Southwest Oncology Group study. J Clin Oncol 1997;15:2414 –2419. 12. Casper ES, Green MR, Kelsen DP, et al. Phase II trial of gemcitabine (2⬘, 2⬘-difluorodeoxycytidine) in patients with

13.

14.

15.

16.

17.

18. 19.

20.

21.

adenocarcinoma of the pancreas. Invest New Drugs 1994;12: 29 –34. Carmichael J, Fink U, Russell RCG, et al. Phase II study of gemcitabine in patients with advanced pancreatic cancer. Br J Cancer 1995;73:101–105. Burriss HA, Moore MJ, Anderson J, et al. Improvements in survival and clinical benefit with gemcitabine as first-line therapy for patients with advanced pancreas cancer: A randomized trial. J Clin Oncol 1997;15:2403–2413. Mohiuddin M, Rosato F, Schuricht A, et al. Carcinoma of the pancreas: The Jefferson experience 1975–1998. Eur J Surg Oncol 1994:20:13–20. Whittington R, Dobelbower RR, Mohiuddin M, et al. Radiotherapy of unresectable pancreatic carcinoma: A six year experience with 104 patients. Int J Radiat Oncol Biol Phys 1981;7:1639 –1644. Andre T, Balosso J, Louvet C, et al. Combined radiotherapy and chemotherapy (cisplatin and 5-fluorouracil) as palliative treatment for localized unresectable or adjuvant treatment for resected pancreatic adenocarcinoma: Results of a feasibility study. Int J Radiat Oncol Biol Phys 2000;46:903–911. Lawrence T. Gemcitabine as a radiation sensitizer. Semin Oncol 1995;22:68 –71. Hoffman JP, McGinn CJ, Szarka C, et al. A phase I study of preoperative gemcitabine (GEM) with radiation therapy (RT) followed by postoperative GEM for patients with localized, resectable pancreatic adenocarcinoma (PAC) (Abstr.). Proc Am Soc Clin Oncol 1998;18:283a. Wolff RA, Evans DB, Gravel DM, et al. Phase I trial of gemcitabine (GEM) combined with radiation (XRT) for the treatment of locally advanced pancreatic adenocarcinoma (Abstr.). Proc Am Soc Clin Oncol 1998;17:283a. Butera J, Malachovsy M, Rathore R, et al. Novel approaches in development for the treatment of pancreatic cancer. Front Bioscience 1998;3:226 –229.