- Email: [email protected]
unresectable pancreatic cancer
Int. J. Radiation Oncology Biol. Phys., Vol. 55, No. 3, pp. 679 – 687, 2003 Copyright © 2003 Elsevier Science Inc. Printed in the USA. All rights reserved 0360-3016/03/$–see front matter
PII S0360-3016(02)03791-4
CLINICAL INVESTIGATION
Pancreas
HYPERFRACTIONATED RADIOTHERAPY AND PACLITAXEL FOR LOCALLY ADVANCED/UNRESECTABLE PANCREATIC CANCER H. ASHAMALLA, M.D., F.C.C.P.,* B. ZAKI, M.D.,* B. MOKHTAR, M.D.,* F. COLELLA, M.D.,† H. SELIM, M.D.,* M. KRISHNAMURTHY, M.D.,† AND P. ROSS, B.S., R.T.T.* Departments of *Radiation Oncology and †Medical Oncology, New York Methodist Hospital, Weill Medical College, Brooklyn, NY Purpose: To determine prospectively the maximal tolerated dose and potential antitumor activity of weekly paclitaxel with concurrent hyperfractionated radiotherapy in patients with locally advanced and/or unresectable pancreatic cancer. Methods and Materials: We embarked on Phase I–II study of hyperfractionated radiotherapy using a concomitant in-field boost to a total dose of 63.80 Gy in 6 weeks at 1.1 Gy/fraction. Paclitaxel was administered weekly on Days 1, 8, 15, 22, 29, and 36 as a 3-h infusion. Paclitaxel doses were escalated from 20 mg/m2/wk to 70 mg/m2/wk. Twenty patients were studied, 14 women and 6 men (mean age 64 years). Some patients presented with one or more symptoms. Obstructive jaundice was the main presenting symptom in 10 patients and epigastric pain in 14. All patients had unresectable histologically proven adenocarcinoma of the pancreas (15 head, 4 body, and 1 tail). Reasons for unresectability were involvement of the portal vein, and/or superior mesenteric artery (n ⴝ 14), paraaortic nodes (n ⴝ 8), and medically inoperable (n ⴝ 1). Fourteen patients underwent a biliary bypass procedure before treatment (four endoscopic stenting, five choledochojejunostomy, and five cholecystojejunostomy). The follow-up period ranged from 14 to 66 months (median 44). Results: The dose-limiting toxicity was observed at 70 mg/m2/wk. Grade IV Radiation Therapy Oncology Group late GI toxicity was seen in 1 patient in the form of duodenal stricture and hemorrhage. Grade II gastrointestinal adverse effects occurred in 13 patients and Grade 3 in 1 patient. No neurologic morbidity was encountered. Eight patients required cytokine support for Grade 2 and 3 neutropenia. The treatment course was delivered within the planned time in 80% of the patients. Complete relief of pain occurred in 10 of 14 patients. The CA 19-9 level was either stable or decreasing in 12 of 15 patients. Of 17 assessable patients, stable disease was seen in 10, regression in 2, a partial response in 3, and a complete response in 2. Conclusion: The use of hyperfractionated radiotherapy to a dose of 63.80 Gy with concomitant weekly paclitaxel is tolerated. The maximal tolerated dose of paclitaxel for this study was 60 mg/m2/wk. The preliminary objective responses denote activity of the regimen. We recommend testing this regimen in larger scale studies. © 2003 Elsevier Science Inc.Elsevier Science Inc. Paclitaxel, Hyperfractionated radiotherapy, Pancreatic cancer, CA 19-9.
INTRODUCTION
disease sites (6 – 8). Paclitaxel radiosensitization occurs through promoting microtubule assembly and arresting the cells at the G2/M phase (9 –13). In addition, it can induce cellular death by promoting apoptosis (10, 12). Several studies (9, 12) have demonstrated the radiosensitizing effects of paclitaxel at low-dose concentrations occurring 7–18 h after paclitaxel exposure (7, 8, 13, 14). We report on a Phase I trial of concomitant hyperfractionated RT combined with paclitaxel as a radiosensitizer in patients with locally advanced and/or unresectable pancreatic cancer.
The estimated number of new cases of pancreatic cancer annually is approximately 30,300, with a reported annual death rate of 14,176 in the United States. The overall 5-year survival rate has remained constant at 5% (1). The patterns of failure of locally advanced or unresectable pancreatic cancer continue to be local or locoregional disease in most cases (2– 4). Several trials have attempted to improve the outcome by using various chemotherapeutic regimens or modified radiotherapy (RT) doses with limited success (1). Hyperfractionated RT offers the advantage of increasing the total tolerable dose with its possible impact on local control without significant increases in late complications (5). Paclitaxel has been used as a radiosensitizer in several
METHODS AND MATERIALS Eligibility The institutional review board approved the study, and all patients gave informed consent. The trial was designed to
Reprint requests to: Hani L. Ashamalla, M.D., F.C.C.P., Department of Radiation Oncology, New York Methodist Hospital, 506 6th St., Brooklyn, NY 11215. Tel: (718) 780-3677; Fax: (718)
780-3688; E-mail: [email protected] Received Feb 19, 2002, and in revised form Jul 29, 2002. Accepted for publication Jul 31, 2002. 679
680
I. J. Radiation Oncology
● Biology ● Physics
Volume 55, Number 3, 2003
Fig. 1. Treatment schema.
establish the maximal tolerated dose (MTD) of paclitaxel administered weekly as a 3-h infusion for 6 consecutive weeks, delivered concurrently with twice-daily irradiation in patients with pancreatic cancer, assessed as locally advanced or unresectable because of adjacent structure involvement or medical inoperability. Eligible patients were ⬎18 years of age, with a Karnofsky performance score ⬎70, who had not received prior RT or chemotherapy for histologically proven pancreatic tumor and had no history of another malignancy in the past 5 years. All patients had acceptable baseline hematologic and renal function (absolute neutrophil count ⬎1800/L, platelet count ⬎100,000/ L, hemoglobin level ⬎10 g/dL, blood urea nitrogen ⬍25 mg/dL and creatinine ⬍1.5 mg/dL). Patients were also required to have adequate hepatic function (bilirubin ⬍2 mg/dL). Patients presenting with obstructive jaundice caused by biliary obstruction from the tumor were included if biliary stenting or surgical bypass had been performed. The pretreatment type and frequency of painkiller use was recorded. The pretreatment assessment included CT scan of the abdomen and pelvis with contrast, as well as determination of CA 19-9 and carcinoembryonic antigen levels. Radiotherapy RT was delivered using 1.1 Gy/fraction twice daily (6-h apart) delivering a total dose of 63.80 Gy in 6 weeks. The initial volume received 44 Gy; it included the tumor as defined on CT scans, as well as areas of potential or proven
nodal involvement, with a 2-cm (planning target volume 1) margin in all directions. The boost volume received 19.80 Gy, including the tumor volume with a 1-cm margin only (planning target volume 2). The upper and lower borders of the boost volume were set to encompass planning target volume 1 with the 95% isodose line. It was given as a concomitant in-field setup, starting daily from the 12th day onward. Figure 1 illustrates the treatment schema. RT was delivered using 6 –15-MV photons using a multiple field arrangement: a four-field technique for the initial volume and a three-field technique for the boost target. Homogeneity within the treatment volume was kept within 5% of the dose at the isocenter volume. Localization and simulation were performed using i.v. pyelography, upper GI series, and CT simulation (AQSIM) for better definition. Doses to critical structures were kept within the known tolerance level: spinal cord ⬍45.00 Gy, two-thirds of both kidneys to ⬍17.00 Gy, and 50% of the liver to ⬍25.00 Gy. Chemotherapy Paclitaxel was delivered during a 3-h i.v. infusion on the first day of every RT week (Monday unless a holiday) for 6 weeks (Fig. 1). Paclitaxel was delivered initially at a dose of 20 mg/m2 to the first 6 patients. The dose was escalated when no Grade IV Radiation Therapy Oncology Group (RTOG) toxicity was encountered in any patient at any level. The MTD was reached if Grade 4 toxicity was encountered in 1 patient. The patients underwent RT within
RT and paclitaxel for pancreatic cancer
Table 1. Number of patients per paclitaxel dose level 2
Paclitaxel dose level (mg/m )
Patients (n)
20 40 50 60 70 Total
6 5 3 4 2 20
2 h of completion of the paclitaxel infusion. Table 1 shows the levels of paclitaxel escalation and the number of patients at each level. The whole regimen was delivered on an outpatient basis. In all patients, premedication with dexamethasone (20 mg i.v.), diphenhydramine (50 mg i.v.), ranitidine (50 mg i.v.), and prochlorperazine (10 mg p.o.) was given 1 h before the start of the infusion. Hematopoietic growth factors were used if the nadir absolute neutrophilic count was ⬍1800/L with no interruptions; treatment was held if the absolute neutrophile count was ⬍700/L. Treatment-related adverse effects were carefully monitored using RTOG/Eastern Cooperative Oncology Group toxicity criteria. Hematologic toxicity was assessed weekly with complete blood cell counts before paclitaxel infusion. Late toxicity was defined as any morbidity at ⱖ3 months. Patients were closely monitored during treatment and thereafter. Measurement of the complete blood count, serum metabolic assay (SMA18), and CA 19-9 were done at each follow-up visit (every 3 months or more frequently if clinically warranted). CT scan of the abdomen was done every 3– 6 months or on a closer schedule if indicated. A complete clinical response required no evidence of tumor on the posttreatment CT scan. A partial response required a 50% reduction of the maximal perpendicular diameters of the tumor. Statistical analysis Although the study was not designed for computation of the survival outcome, Kaplan-Meier survival curves were fitted for death as the end point. The registration date was taken as the starting point of patient time in the analysis. RESULTS Patient characteristics Twenty patients were enrolled in this study. All patients had histologically proven adenocarcinoma of the pancreas. Fifteen tumors were located in the head of the pancreas, four in the body, and one in the tail. Table 2 summarizes the patient characteristics and reasons for unresectability. All patients had locally advanced and/or unresectable disease; 8 had positive preaortic and paraaortic lymph nodes, 14 involvement of the portal vein, superior mesenteric artery and/or vein, and 1 patient was medically inoperable. Fourteen patients had undergone a biliary bypass procedure before treatment (4 endoscopic stenting, 5 choledochojeju-
● H. ASHAMALLA et al.
681
Table 2. Patient characteristics Mean age (n) Sex (M/F) (n) Tumor location Head Body Tail Median KPS Reason for unresectability Mesenteric vessels or portal vein involvement Positive lymph nodes (pre/paraaortic) Medically inoperable Presenting symptoms Abdominal pain Jaundice Both
64 6/14 15 4 1 70 14 8 1 10 6 4
Abbreviations: M/F ⫽ male/female; KPS ⫽ Karnofsky performance score.
nostomy, and 5 cholecystojejunotosomy); one had undergone revision of an endobiliary stent during the course of treatment, and 4 had undergone partial pancreatectomy with documented intraoperative residual disease and/or positive lymph nodes. The median Karnofsky performance score at the time of study entry was 70. The median age was 64 years (range 45– 80). Six patients presented with obstructive jaundice, 10 with abdominal pain, loss of weight and upper GI symptoms, and 4 patients with pain and jaundice. All patients were included in the analysis of tolerance or response. Tolerance to regimen Seventeen patients completed the prescribed course of therapy within the planned time (6 ⫾ 1 week). Three discontinued treatment after one and three courses of paclitaxel. One patient developed septic cholangitis and treatment was stopped after receiving 38.50 Gy and three doses of paclitaxel. Endoscopic retrograde cholangiography revealed no papillary edema that could be related to radiation injury with evidence of obvious inflammation. The event was attributed to a prior stent procedure and was considered unrelated to the treatment. Two patients refused to continue any treatment after 1 week of therapy, with no adverse reactions. No paclitaxel dosage reductions were required because of related toxicity. Eight patients required cytokine support for Grade II and III neutropenia. No platelet transfusion was necessary in any patient. No correlation between hemoglobin toxicity and paclitaxel level was noted. Table 3 illustrates the hematologic toxicities. No Grade 4 hematologic toxicity was observed. Grade III neutropenia was encountered in 4 patients and Grade II in 4. Patients with prior hematologic complications were all normalized. Grade I and II gastrointestinal complications were seen in 18 patients, with only 1 patient needing hospitalization and rehydration for prolonged vomiting (Grade III morbidity). Of all patients with GI complications, only 4 had lower GI symptoms, successfully controlled by phenoxynate and/or
682
I. J. Radiation Oncology
● Biology ● Physics
Volume 55, Number 3, 2003
Table 3. Hematologic toxicity based on RTOG morbidity scales Grade 0 (n)
Grade I (n)
Grade II (n)
Grade III (n)
Grade IV (n)
Paclitaxel dose level (mg/m2)
N
H
P
N
H
P
N
H
P
N
H
P
N
H
P
20 40 50 60 70 Total
0 0 0 0 0 0
0 0 0 0 1 1
5 3 1 4 1 14
4* 0 0 3 1 8
3 2 3 2 0 10
1 2 2 0 1 6
2 0 0 0 0 2
2 2 0 2 0 6
0 0 0 0 0 0
0 0 1 1 0 4
1 1 0 0 1 3
0 0 0 0 0 0
0 0 3 0 0 3
0 0 0 0 0 0
0 0 0 0 0 0
* One patient developed septic cholangitis. Abbreviations: RTOG ⫽ Radiation Therapy Oncology Group; N ⫽ neutrophils; H ⫽ hemoglobin; P ⫽ platelets.
loperamide. When the paclitaxel dose was escalated to 70 mg/m2 Grade IV RTOG toxicity was encountered in 1 of 2 patients at this level; that patient, 4 months after completing treatment, developed duodenal fibrosis, stricture, and bleeding that required hospitalization and attempted surgical repair. That same patient also developed Grade III neuropathy. One patient required parenteral nutrition for episodes of prolonged vomiting at 4 months after treatment. No surgical intervention was necessary. Clinical response Of 17 assessable patients, 6 were alive at last follow-up, with a median follow-up of 44 months (range 14 – 66). The median survival was 10 months. The 1 and 2 year survival rate was 45% and 27%, respectively (Fig. 2). An improved survival rate was observed with increasing paclitaxel dose (p ⫽ 0.19). The median survival for the patients receiving ⱕ40 mg/m2 of weekly paclitaxel was 8 months; the median
survival for those receiving ⱖ50 mg/m2 had not been reached at the time of analysis (Fig. 3). Follow-up CT studies were available for 17 patients. Radiologic stable disease was observed in 10 patients, progressive disease in 2, a partial response in 3 (Fig. 4), and a complete response in 2 (Fig. 5) The latter was seen in patients who received paclitaxel doses of 40 and 70 mg/m2. Three patients were lost to radiologic follow-up. Those who demonstrated radiologic disease progression displayed positive in-field paraaortic nodes and liver metastasis without evidence of local progression. Of those treated with ⬎40 mg/m2 of weekly paclitaxel, 43% had a complete or partial response compared with only 20% of those receiving ⱕ40 mg/m2. The correlation between the paclitaxel level and radiologic response is illustrated in Table 4. An assessment of the patients for various clinical parameters was done at the last follow-up visit. Seven of 20 patients had an increase in body weight, with an average
Fig. 2. Survival curve using Kaplan-Meier estimates. Median survival was 10 months (95% confidence interval).
RT and paclitaxel for pancreatic cancer
● H. ASHAMALLA et al.
683
Fig. 3. Overall survival for patients receiving ⱕ40 mg/m2 of paclitaxel (Group 1) and those receiving ⬎40 mg/m2 (Group 2).
weight gain of 1 kg, and 10 patients had returned to their ideal body weight. Ten of 14 patients who originally presented with variable degrees of pain showed improvement in their pain intensity, and 7 had complete relief without the use of painkillers during throughout the follow-up period. Of all the patients who had undergone prior biliary bypass procedures, only one developed recurrent jaundice and endoscopic stenting was performed again. Posttreatment CA 19-9 values were available for 15 patients and compared with the pretreatment levels. Of those, 6 were stable, 6 had decreased, and 3 had increased. The 2 patients with radiologic evidence of progression were administered other chemotherapy agents for salvage (5-fluorouracil and gemcitabine). One received four cycles of 5-fluorouracil (1 g/mL i.v. infusion for 4 days, repeated every 28 days) and the other received gemcitabine (1 g/mL) weekly for 7 weeks.
improved overall survival compared with the control arm (4). RT consisted of split courses, with 20.0 Gy delivered during 2 weeks and an additional 2-week course of 20.0 Gy after a 2-week rest period. In a study by the Mayo Clinic, of patients receiving 60 Gy, about 67% had symptom improvement, with some improvement in the median survival to 8 months (15), vs. very little improvement after 35 Gy, implying a dose–response equation. Other studies have confirmed this trend (16, 17). In an attempt to enhance locoregional control, two factors were used in our study. Hyperfractionated RT was used to deliver higher RT doses and paclitaxel was used as a sensitizer. The dose selected for this study (63. 80 Gy at 1.10 Gy/fraction twice daily) was extrapolated from the following formula (17): n1d1 a/b ⫹ d2 ⫽ N2d2 a/b ⫹ d1
DISCUSSION About 25% of pancreatic cancer cases are regionally advanced and surgically unresectable (2, 3), without overt or microscopic evidence of metastatic disease at diagnosis. More than three-quarters of the patients will die of progression of the primary pancreatic tumor. Failure in many patients undergoing pancreatectomy usually involves recurrence in the tumor bed with worsening symptoms, mainly pain and jaundice. In a Gastrointestinal Tumor Study Group protocol, patients who received chemoradiotherapy after potentially curative resections had improved disease-free survival and
where n ⫽ number of fractions, d1 ⫽ fraction size of conventional standard treatment (1.80 Gy), d2 ⫽ fraction size of the current study (1.10 Gy), and a/b for late-reacting tissues (e.g., duodenal stroma) ⫽ 3 Gy and a/b for earlyreacting tissues, including duodenal mucosa, ⫽ 10 Gy. Assuming that the conventional dose commonly delivered is 50.40 Gy at 1.80 Gy/fraction, the dose used in this treatment (63.8 Gy at 1.10 Gy/fraction twice daily) should have resulted in a 15% increase in local control with only a 10% increase in late complications. The use of paclitaxel as a radiosensitizer was based on its
684
I. J. Radiation Oncology
● Biology ● Physics
Volume 55, Number 3, 2003
Fig. 4. (a) Pretreatment CT scan. (b) Partial regression of tumor after treatment.
RT and paclitaxel for pancreatic cancer
● H. ASHAMALLA et al.
685
Fig. 5. (a) Pretreatment CT scan. (b) Complete response 2 years after treatment.
function of blocking cells at the G2/M phase (12)—a radiosensitive phase. It was also shown that paclitaxel might act as a radiosensitizer with higher p53 expression (13). Pancreatic cancer has been shown to frequently harbor mutations of p53 (9, 10). Moreover, concomitant paclitaxel and RT enhance apoptotic death at paclitaxel doses that are
much less than its cytotoxic doses (13). We have also shown that the induction of apoptosis was most prominent (34%) at a low concentration of 10 nmol and least (8%) at the higher concentration of 100 nmol (13). The inhibition of culture growth, however, was in the reverse direction. Recently, other in vitro studies have confirmed the higher inhibition of
686
I. J. Radiation Oncology
● Biology ● Physics
Volume 55, Number 3, 2003
Table 4. Radiological response Paclitaxel dose level (mg/m2)
Total (n)
Stable response (n)
Partial response (n)
Complete response (n)
Progressive disease (n)
20 40 50 60 70 Total
6 5 3 4 2 20‡
4 2 2 1 1 10
0 1 1 1 0 3
0 1 0 0 1 2
1* 1† 0 0 0 2
* Progressive paraaortic lymph nodes. † Liver metastasis. ‡ Three patients did not have follow-up CT scan.
culture growth 24 – 48 h after paclitaxel contact, which also adds to the proof of paclitaxel radiosensitization (18). The primary objectives of this trial were to determine the tolerance to the regimen and the MTD of paclitaxel administered weekly during 6 weeks of hyperfractionated RT to the pancreatic fossa. With dose escalation to 70 mg/m2, no significant hematologic or neurologic toxicities were encountered. Cytokine support was used as early as Grade II neutropenia, and subsequently, no major treatment-related interruption was encountered. One patient developed septic cholangitis with normal blood count, this was considered a drug unrelated event. The weekly infusion of paclitaxel was carried out on an outpatient basis. No allergic reactions were seen with the premedication regimen described. Moreover, no major acute GI complications were noted. The GI complications were mainly limited to Grades I and II, except for 2 patients who had Grade III. The MTD was reached at 60 mg/m2, when a Grade IV RTOG GI complication occurred at 70 mg/m2, seen 5 months after treatment in 1 of 2 patients treated at that dose level. The use of a hyperfractionated course of RT to a dose of 63.80 Gy was also well tolerated; 80% of the patients completed the treatment within the planned period. Although the traditional meaning of a concomitant in-field boost is the use of a larger fraction size (1.60 –1.80 Gy) (8) during the boost component, we elected to continue the same (1.1 Gy) fraction size to avoid a higher incidence of late reactions in the duodenum and jejunum. We believe that our ability to deliver a higher radiation dose with a concomitant higher paclitaxel dose with less toxicity was a result of the hyperfractionation and low fraction size. In a recently published similar study using paclitaxel as a radiosensitizer and conventional RT to a total dose of 50 Gy at 1.8 Gy/fraction, Safran and associates (14) reported a significant Grade IV toxicity at a paclitaxel dose of ⬎50 mg/m2 and Grade III nausea and vomiting at 50 mg/m2. The MTD was reached at 60 mg/m2 when 3 patients developed Grade III nausea/anorexia and 2 had Grade III abdominal pain. They recommended 50 mg/m2 as the maximal paclitaxel level for an ongoing Phase II trial. Rich and associates (19) published in abstract form a regimen of 50.40 Gy and
50 mg/mL weekly paclitaxel. The regimen was well tolerated, with a 1-year survival rate of 48%. Although our study was not designed to evaluate the response of pancreatic cancer to this regimen, 5 patients demonstrated radiologic improvement, evident by follow-up CT scans, and 10 had stable disease. The overall local control rate was 88%, approximately 15% greater than published figures (1, 6). This improvement coincides with the predicted biologic hypothesis of our study. Despite the relatively good local control rate, the median survival remained dismal. The cause of death continues to be progressive cancer cachexia in a good percentage of patients. Although the correlation of the objective response with paclitaxel level was difficult to assess, the 2 patients who demonstrated radiologic disease progression had received the lower paclitaxel doses (Table 4). Of the 17 patients with assessable follow-up CT, the overall objective (complete and partial) response rate was 35%. The median survival was 10 months. The 1- and 2-year survival rate was 45% and 27%, respectively. Blackstock et al. (20) demonstrated an 8% objective response rate in a small group of patients with pancreatic cancer who treated with paclitaxel only, not much different from the response rates reported for 5-fluorouracil and gemcitabine (2.5% and 5.4%, respectively) (8), denoting the dismal objective response in the absence of RT. The clinical benefit in this cohort of patients with locally advanced/unresectable pancreatic cancer was also documented through improvement in patient weight on several follow-up visits, as well as a decrease in the pain intensity. Of 14 patients who initially presented with pain, 10 continued to maintain less or no need for analgesics. Because the early and late toxicities encountered in this study were less than the expected ones using the biologic effective dose formula (17), the regimen of hyperfractionation using an in-field boost technique to a dose of 63.80 Gy in 6 weeks concomitant with weekly paclitaxel seems quite tolerable. The MTD of concomitant paclitaxel as a weekly sensitizer was 60 mg/m2 according to our experience. This regimen should be tested further in a Phase II-III trial to establish its benefit and efficacy. Moreover, the use of altered fractionation radiation schemes in pancreatic cancer
RT and paclitaxel for pancreatic cancer
with other active radiosensitizers could be also studied. We believe that the dismal outcome in patients with unresect-
● H. ASHAMALLA et al.
687
able pancreatic cancer warrants additional and continuous efforts to attempt to improve the results.
REFERENCES 1. Jemal A, Thomas A, Murray T, et al. Cancer statistics 2002. CA Cancer J Clin 2002;52:23– 47. 2. Foo ML, Gunderson LL, Nagorney DM, et al. Patterns of failure in grossly resected pancreatic ductal adenocarcinoma treated with adjuvant irradiation ⫾ 5-flourouracil. Int J Radiat Oncol Biol Phys 1993;26:483– 489. 3. Tepper, J, Nardi G, Susi H. Carcinoma of the pancreas: Review of the MGH experience from 1963–1973. Cancer 1976;37:1519 –1524. 4. Griffin JF, Smalley SR, Jewell W, et al. Patterns of failure after curative resection of pancreatic carcinoma. Cancer 1990; 66:56 – 61. 5. Garden AS, Morrison WH, Ang KK, Peters LJ. Hyperfractionated radiation in the treatment of squamous cell carcinomas of the head and neck: A comparison of two fractionated schedules. Int J Radiat Oncol Biol Phys 1995;31:493–502. 6. Hoffman JP, Weese JL, Lipsitz S, et al. Preoperative chemoradiation for patients with resectable pancreatic adenocarcinoma: An Eastern Cooperative Oncology Group (ECOG) phase II study [Abstract]. Proc Am Clin Oncol 1995;14:201. 7. Glantz M, Choy H, Kearns CM, et al. Phase I study of weekly outpatient paclitaxel and concurrent cranial irradiation in adults with astrocytomas. J Clin Oncol 1996;14:600 – 609. 8. Choy H, Akerley W, Safran H, et al. Phase I trial of outpatient weekly paclitaxel and concurrent radiation therapy for advanced non-small-cell lung cancer. J Clin Oncol 1994;12: 2682–2686. 9. Liebmann J, Cook JA, Fisher J, et al. In vitro studies of Taxol as a radiation sensitizer in human tumor cells. J Natl Cancer Inst 1994;86:441– 446. 10. Choy H, Rodriguez FF, Koester S, et al. Investigation of Taxol as potential radiation sensitizer. Cancer 1993;71:3774 –3778. 11. Nigro JM, Baker SJ, Preisinger AC, et al. Mutation in the p53
12.
13. 14.
15.
16.
17. 18.
19.
20.
gene occur in diverse human tumor types. Nature 1989;342: 705–708. Yang SJ, Ashamalla H, Youssef A. Mitotic arrest: Cell death and increased p53 immunofluorescence in Taxol treated cells. Proc ASCO 1995;14:164. Yang SJ, Ashamalla H. Taxol cytotoxicity and its relation to exposure time. Proc ASCO 1996;15:338. Safran H, King TP, 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. Haslam JB, Cavanaugh PJ, Stroup SL. Radiation therapy in the treatment of irresectable adenocarcinoma of the pancreas. Cancer 1973;32:1341–1345. Dobelbower RR, Bargelt BB, Strubler KA, et al. Precision radiotherapy for cancer of the pancreas: Techniques and results. Int J Radiat Oncol Biol Phys 1980;6:1127–1233. Fowler JF. Brief summary of radiobiological principle in fractionated radiotherapy. Semin Radiat Oncol 1992;2:16 –21. Mason KA, Hunter NR, Milas L, et al. Enhancement of radiation response of mouse mammary tumors in-vivo. Proceedings of International Congress on Breast Disease, M. D. Anderson, OR4, 1996. Abstract, p. 204. Rich T, Harris J, Abrams R, et al. A Phase II study of external irradiation and weekly paclitaxel for non-metastatic, unresectable pancreatic cancer: A preliminary report of RTOG protocol 98-12 (Abstract). Int J Radiat Oncol Biol Phys 2001;51: 29 –30. Blackstock AW, Lesser GJ, Fletcher-Steede J, et al. Phase I study of twice-weekly gemcitabine and concurrent thoracic radiation for patients with locally advanced non–small-cell lung cancer. Int J Radiat Oncol Biol Phys 2001;51:1281– 1289.
PII S0360-3016(02)03791-4
CLINICAL INVESTIGATION
Pancreas
HYPERFRACTIONATED RADIOTHERAPY AND PACLITAXEL FOR LOCALLY ADVANCED/UNRESECTABLE PANCREATIC CANCER H. ASHAMALLA, M.D., F.C.C.P.,* B. ZAKI, M.D.,* B. MOKHTAR, M.D.,* F. COLELLA, M.D.,† H. SELIM, M.D.,* M. KRISHNAMURTHY, M.D.,† AND P. ROSS, B.S., R.T.T.* Departments of *Radiation Oncology and †Medical Oncology, New York Methodist Hospital, Weill Medical College, Brooklyn, NY Purpose: To determine prospectively the maximal tolerated dose and potential antitumor activity of weekly paclitaxel with concurrent hyperfractionated radiotherapy in patients with locally advanced and/or unresectable pancreatic cancer. Methods and Materials: We embarked on Phase I–II study of hyperfractionated radiotherapy using a concomitant in-field boost to a total dose of 63.80 Gy in 6 weeks at 1.1 Gy/fraction. Paclitaxel was administered weekly on Days 1, 8, 15, 22, 29, and 36 as a 3-h infusion. Paclitaxel doses were escalated from 20 mg/m2/wk to 70 mg/m2/wk. Twenty patients were studied, 14 women and 6 men (mean age 64 years). Some patients presented with one or more symptoms. Obstructive jaundice was the main presenting symptom in 10 patients and epigastric pain in 14. All patients had unresectable histologically proven adenocarcinoma of the pancreas (15 head, 4 body, and 1 tail). Reasons for unresectability were involvement of the portal vein, and/or superior mesenteric artery (n ⴝ 14), paraaortic nodes (n ⴝ 8), and medically inoperable (n ⴝ 1). Fourteen patients underwent a biliary bypass procedure before treatment (four endoscopic stenting, five choledochojejunostomy, and five cholecystojejunostomy). The follow-up period ranged from 14 to 66 months (median 44). Results: The dose-limiting toxicity was observed at 70 mg/m2/wk. Grade IV Radiation Therapy Oncology Group late GI toxicity was seen in 1 patient in the form of duodenal stricture and hemorrhage. Grade II gastrointestinal adverse effects occurred in 13 patients and Grade 3 in 1 patient. No neurologic morbidity was encountered. Eight patients required cytokine support for Grade 2 and 3 neutropenia. The treatment course was delivered within the planned time in 80% of the patients. Complete relief of pain occurred in 10 of 14 patients. The CA 19-9 level was either stable or decreasing in 12 of 15 patients. Of 17 assessable patients, stable disease was seen in 10, regression in 2, a partial response in 3, and a complete response in 2. Conclusion: The use of hyperfractionated radiotherapy to a dose of 63.80 Gy with concomitant weekly paclitaxel is tolerated. The maximal tolerated dose of paclitaxel for this study was 60 mg/m2/wk. The preliminary objective responses denote activity of the regimen. We recommend testing this regimen in larger scale studies. © 2003 Elsevier Science Inc.Elsevier Science Inc. Paclitaxel, Hyperfractionated radiotherapy, Pancreatic cancer, CA 19-9.
INTRODUCTION
disease sites (6 – 8). Paclitaxel radiosensitization occurs through promoting microtubule assembly and arresting the cells at the G2/M phase (9 –13). In addition, it can induce cellular death by promoting apoptosis (10, 12). Several studies (9, 12) have demonstrated the radiosensitizing effects of paclitaxel at low-dose concentrations occurring 7–18 h after paclitaxel exposure (7, 8, 13, 14). We report on a Phase I trial of concomitant hyperfractionated RT combined with paclitaxel as a radiosensitizer in patients with locally advanced and/or unresectable pancreatic cancer.
The estimated number of new cases of pancreatic cancer annually is approximately 30,300, with a reported annual death rate of 14,176 in the United States. The overall 5-year survival rate has remained constant at 5% (1). The patterns of failure of locally advanced or unresectable pancreatic cancer continue to be local or locoregional disease in most cases (2– 4). Several trials have attempted to improve the outcome by using various chemotherapeutic regimens or modified radiotherapy (RT) doses with limited success (1). Hyperfractionated RT offers the advantage of increasing the total tolerable dose with its possible impact on local control without significant increases in late complications (5). Paclitaxel has been used as a radiosensitizer in several
METHODS AND MATERIALS Eligibility The institutional review board approved the study, and all patients gave informed consent. The trial was designed to
Reprint requests to: Hani L. Ashamalla, M.D., F.C.C.P., Department of Radiation Oncology, New York Methodist Hospital, 506 6th St., Brooklyn, NY 11215. Tel: (718) 780-3677; Fax: (718)
780-3688; E-mail: [email protected] Received Feb 19, 2002, and in revised form Jul 29, 2002. Accepted for publication Jul 31, 2002. 679
680
I. J. Radiation Oncology
● Biology ● Physics
Volume 55, Number 3, 2003
Fig. 1. Treatment schema.
establish the maximal tolerated dose (MTD) of paclitaxel administered weekly as a 3-h infusion for 6 consecutive weeks, delivered concurrently with twice-daily irradiation in patients with pancreatic cancer, assessed as locally advanced or unresectable because of adjacent structure involvement or medical inoperability. Eligible patients were ⬎18 years of age, with a Karnofsky performance score ⬎70, who had not received prior RT or chemotherapy for histologically proven pancreatic tumor and had no history of another malignancy in the past 5 years. All patients had acceptable baseline hematologic and renal function (absolute neutrophil count ⬎1800/L, platelet count ⬎100,000/ L, hemoglobin level ⬎10 g/dL, blood urea nitrogen ⬍25 mg/dL and creatinine ⬍1.5 mg/dL). Patients were also required to have adequate hepatic function (bilirubin ⬍2 mg/dL). Patients presenting with obstructive jaundice caused by biliary obstruction from the tumor were included if biliary stenting or surgical bypass had been performed. The pretreatment type and frequency of painkiller use was recorded. The pretreatment assessment included CT scan of the abdomen and pelvis with contrast, as well as determination of CA 19-9 and carcinoembryonic antigen levels. Radiotherapy RT was delivered using 1.1 Gy/fraction twice daily (6-h apart) delivering a total dose of 63.80 Gy in 6 weeks. The initial volume received 44 Gy; it included the tumor as defined on CT scans, as well as areas of potential or proven
nodal involvement, with a 2-cm (planning target volume 1) margin in all directions. The boost volume received 19.80 Gy, including the tumor volume with a 1-cm margin only (planning target volume 2). The upper and lower borders of the boost volume were set to encompass planning target volume 1 with the 95% isodose line. It was given as a concomitant in-field setup, starting daily from the 12th day onward. Figure 1 illustrates the treatment schema. RT was delivered using 6 –15-MV photons using a multiple field arrangement: a four-field technique for the initial volume and a three-field technique for the boost target. Homogeneity within the treatment volume was kept within 5% of the dose at the isocenter volume. Localization and simulation were performed using i.v. pyelography, upper GI series, and CT simulation (AQSIM) for better definition. Doses to critical structures were kept within the known tolerance level: spinal cord ⬍45.00 Gy, two-thirds of both kidneys to ⬍17.00 Gy, and 50% of the liver to ⬍25.00 Gy. Chemotherapy Paclitaxel was delivered during a 3-h i.v. infusion on the first day of every RT week (Monday unless a holiday) for 6 weeks (Fig. 1). Paclitaxel was delivered initially at a dose of 20 mg/m2 to the first 6 patients. The dose was escalated when no Grade IV Radiation Therapy Oncology Group (RTOG) toxicity was encountered in any patient at any level. The MTD was reached if Grade 4 toxicity was encountered in 1 patient. The patients underwent RT within
RT and paclitaxel for pancreatic cancer
Table 1. Number of patients per paclitaxel dose level 2
Paclitaxel dose level (mg/m )
Patients (n)
20 40 50 60 70 Total
6 5 3 4 2 20
2 h of completion of the paclitaxel infusion. Table 1 shows the levels of paclitaxel escalation and the number of patients at each level. The whole regimen was delivered on an outpatient basis. In all patients, premedication with dexamethasone (20 mg i.v.), diphenhydramine (50 mg i.v.), ranitidine (50 mg i.v.), and prochlorperazine (10 mg p.o.) was given 1 h before the start of the infusion. Hematopoietic growth factors were used if the nadir absolute neutrophilic count was ⬍1800/L with no interruptions; treatment was held if the absolute neutrophile count was ⬍700/L. Treatment-related adverse effects were carefully monitored using RTOG/Eastern Cooperative Oncology Group toxicity criteria. Hematologic toxicity was assessed weekly with complete blood cell counts before paclitaxel infusion. Late toxicity was defined as any morbidity at ⱖ3 months. Patients were closely monitored during treatment and thereafter. Measurement of the complete blood count, serum metabolic assay (SMA18), and CA 19-9 were done at each follow-up visit (every 3 months or more frequently if clinically warranted). CT scan of the abdomen was done every 3– 6 months or on a closer schedule if indicated. A complete clinical response required no evidence of tumor on the posttreatment CT scan. A partial response required a 50% reduction of the maximal perpendicular diameters of the tumor. Statistical analysis Although the study was not designed for computation of the survival outcome, Kaplan-Meier survival curves were fitted for death as the end point. The registration date was taken as the starting point of patient time in the analysis. RESULTS Patient characteristics Twenty patients were enrolled in this study. All patients had histologically proven adenocarcinoma of the pancreas. Fifteen tumors were located in the head of the pancreas, four in the body, and one in the tail. Table 2 summarizes the patient characteristics and reasons for unresectability. All patients had locally advanced and/or unresectable disease; 8 had positive preaortic and paraaortic lymph nodes, 14 involvement of the portal vein, superior mesenteric artery and/or vein, and 1 patient was medically inoperable. Fourteen patients had undergone a biliary bypass procedure before treatment (4 endoscopic stenting, 5 choledochojeju-
● H. ASHAMALLA et al.
681
Table 2. Patient characteristics Mean age (n) Sex (M/F) (n) Tumor location Head Body Tail Median KPS Reason for unresectability Mesenteric vessels or portal vein involvement Positive lymph nodes (pre/paraaortic) Medically inoperable Presenting symptoms Abdominal pain Jaundice Both
64 6/14 15 4 1 70 14 8 1 10 6 4
Abbreviations: M/F ⫽ male/female; KPS ⫽ Karnofsky performance score.
nostomy, and 5 cholecystojejunotosomy); one had undergone revision of an endobiliary stent during the course of treatment, and 4 had undergone partial pancreatectomy with documented intraoperative residual disease and/or positive lymph nodes. The median Karnofsky performance score at the time of study entry was 70. The median age was 64 years (range 45– 80). Six patients presented with obstructive jaundice, 10 with abdominal pain, loss of weight and upper GI symptoms, and 4 patients with pain and jaundice. All patients were included in the analysis of tolerance or response. Tolerance to regimen Seventeen patients completed the prescribed course of therapy within the planned time (6 ⫾ 1 week). Three discontinued treatment after one and three courses of paclitaxel. One patient developed septic cholangitis and treatment was stopped after receiving 38.50 Gy and three doses of paclitaxel. Endoscopic retrograde cholangiography revealed no papillary edema that could be related to radiation injury with evidence of obvious inflammation. The event was attributed to a prior stent procedure and was considered unrelated to the treatment. Two patients refused to continue any treatment after 1 week of therapy, with no adverse reactions. No paclitaxel dosage reductions were required because of related toxicity. Eight patients required cytokine support for Grade II and III neutropenia. No platelet transfusion was necessary in any patient. No correlation between hemoglobin toxicity and paclitaxel level was noted. Table 3 illustrates the hematologic toxicities. No Grade 4 hematologic toxicity was observed. Grade III neutropenia was encountered in 4 patients and Grade II in 4. Patients with prior hematologic complications were all normalized. Grade I and II gastrointestinal complications were seen in 18 patients, with only 1 patient needing hospitalization and rehydration for prolonged vomiting (Grade III morbidity). Of all patients with GI complications, only 4 had lower GI symptoms, successfully controlled by phenoxynate and/or
682
I. J. Radiation Oncology
● Biology ● Physics
Volume 55, Number 3, 2003
Table 3. Hematologic toxicity based on RTOG morbidity scales Grade 0 (n)
Grade I (n)
Grade II (n)
Grade III (n)
Grade IV (n)
Paclitaxel dose level (mg/m2)
N
H
P
N
H
P
N
H
P
N
H
P
N
H
P
20 40 50 60 70 Total
0 0 0 0 0 0
0 0 0 0 1 1
5 3 1 4 1 14
4* 0 0 3 1 8
3 2 3 2 0 10
1 2 2 0 1 6
2 0 0 0 0 2
2 2 0 2 0 6
0 0 0 0 0 0
0 0 1 1 0 4
1 1 0 0 1 3
0 0 0 0 0 0
0 0 3 0 0 3
0 0 0 0 0 0
0 0 0 0 0 0
* One patient developed septic cholangitis. Abbreviations: RTOG ⫽ Radiation Therapy Oncology Group; N ⫽ neutrophils; H ⫽ hemoglobin; P ⫽ platelets.
loperamide. When the paclitaxel dose was escalated to 70 mg/m2 Grade IV RTOG toxicity was encountered in 1 of 2 patients at this level; that patient, 4 months after completing treatment, developed duodenal fibrosis, stricture, and bleeding that required hospitalization and attempted surgical repair. That same patient also developed Grade III neuropathy. One patient required parenteral nutrition for episodes of prolonged vomiting at 4 months after treatment. No surgical intervention was necessary. Clinical response Of 17 assessable patients, 6 were alive at last follow-up, with a median follow-up of 44 months (range 14 – 66). The median survival was 10 months. The 1 and 2 year survival rate was 45% and 27%, respectively (Fig. 2). An improved survival rate was observed with increasing paclitaxel dose (p ⫽ 0.19). The median survival for the patients receiving ⱕ40 mg/m2 of weekly paclitaxel was 8 months; the median
survival for those receiving ⱖ50 mg/m2 had not been reached at the time of analysis (Fig. 3). Follow-up CT studies were available for 17 patients. Radiologic stable disease was observed in 10 patients, progressive disease in 2, a partial response in 3 (Fig. 4), and a complete response in 2 (Fig. 5) The latter was seen in patients who received paclitaxel doses of 40 and 70 mg/m2. Three patients were lost to radiologic follow-up. Those who demonstrated radiologic disease progression displayed positive in-field paraaortic nodes and liver metastasis without evidence of local progression. Of those treated with ⬎40 mg/m2 of weekly paclitaxel, 43% had a complete or partial response compared with only 20% of those receiving ⱕ40 mg/m2. The correlation between the paclitaxel level and radiologic response is illustrated in Table 4. An assessment of the patients for various clinical parameters was done at the last follow-up visit. Seven of 20 patients had an increase in body weight, with an average
Fig. 2. Survival curve using Kaplan-Meier estimates. Median survival was 10 months (95% confidence interval).
RT and paclitaxel for pancreatic cancer
● H. ASHAMALLA et al.
683
Fig. 3. Overall survival for patients receiving ⱕ40 mg/m2 of paclitaxel (Group 1) and those receiving ⬎40 mg/m2 (Group 2).
weight gain of 1 kg, and 10 patients had returned to their ideal body weight. Ten of 14 patients who originally presented with variable degrees of pain showed improvement in their pain intensity, and 7 had complete relief without the use of painkillers during throughout the follow-up period. Of all the patients who had undergone prior biliary bypass procedures, only one developed recurrent jaundice and endoscopic stenting was performed again. Posttreatment CA 19-9 values were available for 15 patients and compared with the pretreatment levels. Of those, 6 were stable, 6 had decreased, and 3 had increased. The 2 patients with radiologic evidence of progression were administered other chemotherapy agents for salvage (5-fluorouracil and gemcitabine). One received four cycles of 5-fluorouracil (1 g/mL i.v. infusion for 4 days, repeated every 28 days) and the other received gemcitabine (1 g/mL) weekly for 7 weeks.
improved overall survival compared with the control arm (4). RT consisted of split courses, with 20.0 Gy delivered during 2 weeks and an additional 2-week course of 20.0 Gy after a 2-week rest period. In a study by the Mayo Clinic, of patients receiving 60 Gy, about 67% had symptom improvement, with some improvement in the median survival to 8 months (15), vs. very little improvement after 35 Gy, implying a dose–response equation. Other studies have confirmed this trend (16, 17). In an attempt to enhance locoregional control, two factors were used in our study. Hyperfractionated RT was used to deliver higher RT doses and paclitaxel was used as a sensitizer. The dose selected for this study (63. 80 Gy at 1.10 Gy/fraction twice daily) was extrapolated from the following formula (17): n1d1 a/b ⫹ d2 ⫽ N2d2 a/b ⫹ d1
DISCUSSION About 25% of pancreatic cancer cases are regionally advanced and surgically unresectable (2, 3), without overt or microscopic evidence of metastatic disease at diagnosis. More than three-quarters of the patients will die of progression of the primary pancreatic tumor. Failure in many patients undergoing pancreatectomy usually involves recurrence in the tumor bed with worsening symptoms, mainly pain and jaundice. In a Gastrointestinal Tumor Study Group protocol, patients who received chemoradiotherapy after potentially curative resections had improved disease-free survival and
where n ⫽ number of fractions, d1 ⫽ fraction size of conventional standard treatment (1.80 Gy), d2 ⫽ fraction size of the current study (1.10 Gy), and a/b for late-reacting tissues (e.g., duodenal stroma) ⫽ 3 Gy and a/b for earlyreacting tissues, including duodenal mucosa, ⫽ 10 Gy. Assuming that the conventional dose commonly delivered is 50.40 Gy at 1.80 Gy/fraction, the dose used in this treatment (63.8 Gy at 1.10 Gy/fraction twice daily) should have resulted in a 15% increase in local control with only a 10% increase in late complications. The use of paclitaxel as a radiosensitizer was based on its
684
I. J. Radiation Oncology
● Biology ● Physics
Volume 55, Number 3, 2003
Fig. 4. (a) Pretreatment CT scan. (b) Partial regression of tumor after treatment.
RT and paclitaxel for pancreatic cancer
● H. ASHAMALLA et al.
685
Fig. 5. (a) Pretreatment CT scan. (b) Complete response 2 years after treatment.
function of blocking cells at the G2/M phase (12)—a radiosensitive phase. It was also shown that paclitaxel might act as a radiosensitizer with higher p53 expression (13). Pancreatic cancer has been shown to frequently harbor mutations of p53 (9, 10). Moreover, concomitant paclitaxel and RT enhance apoptotic death at paclitaxel doses that are
much less than its cytotoxic doses (13). We have also shown that the induction of apoptosis was most prominent (34%) at a low concentration of 10 nmol and least (8%) at the higher concentration of 100 nmol (13). The inhibition of culture growth, however, was in the reverse direction. Recently, other in vitro studies have confirmed the higher inhibition of
686
I. J. Radiation Oncology
● Biology ● Physics
Volume 55, Number 3, 2003
Table 4. Radiological response Paclitaxel dose level (mg/m2)
Total (n)
Stable response (n)
Partial response (n)
Complete response (n)
Progressive disease (n)
20 40 50 60 70 Total
6 5 3 4 2 20‡
4 2 2 1 1 10
0 1 1 1 0 3
0 1 0 0 1 2
1* 1† 0 0 0 2
* Progressive paraaortic lymph nodes. † Liver metastasis. ‡ Three patients did not have follow-up CT scan.
culture growth 24 – 48 h after paclitaxel contact, which also adds to the proof of paclitaxel radiosensitization (18). The primary objectives of this trial were to determine the tolerance to the regimen and the MTD of paclitaxel administered weekly during 6 weeks of hyperfractionated RT to the pancreatic fossa. With dose escalation to 70 mg/m2, no significant hematologic or neurologic toxicities were encountered. Cytokine support was used as early as Grade II neutropenia, and subsequently, no major treatment-related interruption was encountered. One patient developed septic cholangitis with normal blood count, this was considered a drug unrelated event. The weekly infusion of paclitaxel was carried out on an outpatient basis. No allergic reactions were seen with the premedication regimen described. Moreover, no major acute GI complications were noted. The GI complications were mainly limited to Grades I and II, except for 2 patients who had Grade III. The MTD was reached at 60 mg/m2, when a Grade IV RTOG GI complication occurred at 70 mg/m2, seen 5 months after treatment in 1 of 2 patients treated at that dose level. The use of a hyperfractionated course of RT to a dose of 63.80 Gy was also well tolerated; 80% of the patients completed the treatment within the planned period. Although the traditional meaning of a concomitant in-field boost is the use of a larger fraction size (1.60 –1.80 Gy) (8) during the boost component, we elected to continue the same (1.1 Gy) fraction size to avoid a higher incidence of late reactions in the duodenum and jejunum. We believe that our ability to deliver a higher radiation dose with a concomitant higher paclitaxel dose with less toxicity was a result of the hyperfractionation and low fraction size. In a recently published similar study using paclitaxel as a radiosensitizer and conventional RT to a total dose of 50 Gy at 1.8 Gy/fraction, Safran and associates (14) reported a significant Grade IV toxicity at a paclitaxel dose of ⬎50 mg/m2 and Grade III nausea and vomiting at 50 mg/m2. The MTD was reached at 60 mg/m2 when 3 patients developed Grade III nausea/anorexia and 2 had Grade III abdominal pain. They recommended 50 mg/m2 as the maximal paclitaxel level for an ongoing Phase II trial. Rich and associates (19) published in abstract form a regimen of 50.40 Gy and
50 mg/mL weekly paclitaxel. The regimen was well tolerated, with a 1-year survival rate of 48%. Although our study was not designed to evaluate the response of pancreatic cancer to this regimen, 5 patients demonstrated radiologic improvement, evident by follow-up CT scans, and 10 had stable disease. The overall local control rate was 88%, approximately 15% greater than published figures (1, 6). This improvement coincides with the predicted biologic hypothesis of our study. Despite the relatively good local control rate, the median survival remained dismal. The cause of death continues to be progressive cancer cachexia in a good percentage of patients. Although the correlation of the objective response with paclitaxel level was difficult to assess, the 2 patients who demonstrated radiologic disease progression had received the lower paclitaxel doses (Table 4). Of the 17 patients with assessable follow-up CT, the overall objective (complete and partial) response rate was 35%. The median survival was 10 months. The 1- and 2-year survival rate was 45% and 27%, respectively. Blackstock et al. (20) demonstrated an 8% objective response rate in a small group of patients with pancreatic cancer who treated with paclitaxel only, not much different from the response rates reported for 5-fluorouracil and gemcitabine (2.5% and 5.4%, respectively) (8), denoting the dismal objective response in the absence of RT. The clinical benefit in this cohort of patients with locally advanced/unresectable pancreatic cancer was also documented through improvement in patient weight on several follow-up visits, as well as a decrease in the pain intensity. Of 14 patients who initially presented with pain, 10 continued to maintain less or no need for analgesics. Because the early and late toxicities encountered in this study were less than the expected ones using the biologic effective dose formula (17), the regimen of hyperfractionation using an in-field boost technique to a dose of 63.80 Gy in 6 weeks concomitant with weekly paclitaxel seems quite tolerable. The MTD of concomitant paclitaxel as a weekly sensitizer was 60 mg/m2 according to our experience. This regimen should be tested further in a Phase II-III trial to establish its benefit and efficacy. Moreover, the use of altered fractionation radiation schemes in pancreatic cancer
RT and paclitaxel for pancreatic cancer
with other active radiosensitizers could be also studied. We believe that the dismal outcome in patients with unresect-
● H. ASHAMALLA et al.
687
able pancreatic cancer warrants additional and continuous efforts to attempt to improve the results.
REFERENCES 1. Jemal A, Thomas A, Murray T, et al. Cancer statistics 2002. CA Cancer J Clin 2002;52:23– 47. 2. Foo ML, Gunderson LL, Nagorney DM, et al. Patterns of failure in grossly resected pancreatic ductal adenocarcinoma treated with adjuvant irradiation ⫾ 5-flourouracil. Int J Radiat Oncol Biol Phys 1993;26:483– 489. 3. Tepper, J, Nardi G, Susi H. Carcinoma of the pancreas: Review of the MGH experience from 1963–1973. Cancer 1976;37:1519 –1524. 4. Griffin JF, Smalley SR, Jewell W, et al. Patterns of failure after curative resection of pancreatic carcinoma. Cancer 1990; 66:56 – 61. 5. Garden AS, Morrison WH, Ang KK, Peters LJ. Hyperfractionated radiation in the treatment of squamous cell carcinomas of the head and neck: A comparison of two fractionated schedules. Int J Radiat Oncol Biol Phys 1995;31:493–502. 6. Hoffman JP, Weese JL, Lipsitz S, et al. Preoperative chemoradiation for patients with resectable pancreatic adenocarcinoma: An Eastern Cooperative Oncology Group (ECOG) phase II study [Abstract]. Proc Am Clin Oncol 1995;14:201. 7. Glantz M, Choy H, Kearns CM, et al. Phase I study of weekly outpatient paclitaxel and concurrent cranial irradiation in adults with astrocytomas. J Clin Oncol 1996;14:600 – 609. 8. Choy H, Akerley W, Safran H, et al. Phase I trial of outpatient weekly paclitaxel and concurrent radiation therapy for advanced non-small-cell lung cancer. J Clin Oncol 1994;12: 2682–2686. 9. Liebmann J, Cook JA, Fisher J, et al. In vitro studies of Taxol as a radiation sensitizer in human tumor cells. J Natl Cancer Inst 1994;86:441– 446. 10. Choy H, Rodriguez FF, Koester S, et al. Investigation of Taxol as potential radiation sensitizer. Cancer 1993;71:3774 –3778. 11. Nigro JM, Baker SJ, Preisinger AC, et al. Mutation in the p53
12.
13. 14.
15.
16.
17. 18.
19.
20.
gene occur in diverse human tumor types. Nature 1989;342: 705–708. Yang SJ, Ashamalla H, Youssef A. Mitotic arrest: Cell death and increased p53 immunofluorescence in Taxol treated cells. Proc ASCO 1995;14:164. Yang SJ, Ashamalla H. Taxol cytotoxicity and its relation to exposure time. Proc ASCO 1996;15:338. Safran H, King TP, 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. Haslam JB, Cavanaugh PJ, Stroup SL. Radiation therapy in the treatment of irresectable adenocarcinoma of the pancreas. Cancer 1973;32:1341–1345. Dobelbower RR, Bargelt BB, Strubler KA, et al. Precision radiotherapy for cancer of the pancreas: Techniques and results. Int J Radiat Oncol Biol Phys 1980;6:1127–1233. Fowler JF. Brief summary of radiobiological principle in fractionated radiotherapy. Semin Radiat Oncol 1992;2:16 –21. Mason KA, Hunter NR, Milas L, et al. Enhancement of radiation response of mouse mammary tumors in-vivo. Proceedings of International Congress on Breast Disease, M. D. Anderson, OR4, 1996. Abstract, p. 204. Rich T, Harris J, Abrams R, et al. A Phase II study of external irradiation and weekly paclitaxel for non-metastatic, unresectable pancreatic cancer: A preliminary report of RTOG protocol 98-12 (Abstract). Int J Radiat Oncol Biol Phys 2001;51: 29 –30. Blackstock AW, Lesser GJ, Fletcher-Steede J, et al. Phase I study of twice-weekly gemcitabine and concurrent thoracic radiation for patients with locally advanced non–small-cell lung cancer. Int J Radiat Oncol Biol Phys 2001;51:1281– 1289.