1233

Evaluation of 96-Hour Infusion Fluorouracil plus Cisplatin in Combination with Alpha Interferon for Patients with Advanced Squamous Cell Carcinoma of the Head and Neck A Southwest Oncology Group Study

Maha Hussain, M.D.,*.t lacqueline Benedetti, Ph.D.,S Roy E. Smith, M.D.,€j Gladys I. Rodriguez, M.D.,(I David Schuller, M.D.,S and John Ensley, M.D.t

Background. Recurrent cancer of the head and neck after primary therapy is almost always fatal. The combi- nation of 5-fluorouracil(5-FU) and cisplatin is considered the best available therapy but complete response rates re- main too low to affect survival. This study was designed to evaluate the complete response rate and toxicity of 5- FU, cisplatin, and alpha-interferon (a-IFN) in patients with recurrent or metastatic squamous cell carcinoma of the head and neck (SCCHN).

Methods. Fifty eligible patients with recurrent or metastatic SCCHN and no prior chemotherapy (40 men, 10 women; age range, 26-77 years; median, 59 years; 82% white; 88% had prior surgery and 92% had prior radia- tion therapy) were treated every 21 days with 96-hour in- fusion of 5-FU 1000 mg/m2/day; cisplatin 100 mg/m*, day 1; and a-IFN 5 X lo6 units/day, days 1-4.

Results. One hundred fifty-seven courses of chemo- therapy were administered, with a median of three courses. Thirty-seven patients experienced Grade 3 or 4 toxicity. Of the 17 patients with Grade 4 toxicity; 12 had hematologic toxicity, 3 stomatitis, and 2 vomiting. Two additional patients died of myelosuppression-related sep-

From the *Veterans Administration Medical Center, Allen Park, Michigan; tWayne State University, Detroit, Michigan; $Southwest Oncology Group Statistical Center, Seattle, Washington; jjOhio State University Health Center, Columbus, Ohio; ((University of Texas Health Science Center at San Antonio, San Antonio, Texas.

Supported in part by the following PHS Cooperative Agreement grant numbers awarded by the National Cancer Institute, DHHS: CA38926, CA32102, CA04920, CA42028, CA22433, CA12644, CA46113, CA35261, CA35176, CA58658, CA27057, CA37981, CA45450, CA04919, CA35090, CA16385, CA16385, CA58686, CA45560, CA20319, CA58861.

Address for reprints: Janet Groff, Southwest Oncology Group (SWOG-9062), Operations Office, 14980 Omicron Drive, San Anto- nio, Texas 78245-3217.

Received May 15,1995; accepted June 1, 1995.

sis. Of the 50 patients, 3 (6%) achieved a complete re- sponse, five (10%) had a partial response, 3 (6%) had un- confirmed response (1 complete and 2 partial), 10 (20%) had stable disease, 17 (34%) progressed, and 12 (24%) were considered nonresponders owing to early death (6) or inadequate assessment (6). The median survival was 5 months.

Conclusion. The complete response rate of patients with recurrent or metastatic SCCHN treated with 5-FU, cisplatin, and a-IFN does not appear to be superior to that observed for 5-FU and cisplatin. Alpha-interferon ap- pears to augment hematologic and gastrointestinal toxic- ities associated with this combination. Cancer 1995;76: 1233-7.

Key words: chemotherapy, recurrent head and neck can- cer, a-interferon, biomodulation.

Recurrent squamous cell carcinoma of the head and neck (SCCHN) is a fatal disease that responds poorly to chemotherapy, A variety of single agents can induce brief partial responses. Of these, 5-fluorouracil (5-FU) and cisplatin have been two of the most active, each with a single-agent response rate of 3070.' This re- sponse has been improved upon by combining the two drugs, resulting in an overall response rate ranging from 11% to 75%.' In a recent Southwest Oncology Group (SWOG) Phase I11 trial, the combination of 5-FU plus cisplatin appeared superior to single-agent methotrex- ate and to the combination of 5-FU plus carboplatin, with overall response rates of 30%, 11%, and 18%, re- ~pectively.~ However, combination chtmotherapy-in- duced complete responses in patients with recurrent and/or metastatic disease remain too low to affect sur-

1234 CANCER October 1,1995, Volume 76, No. 7

vival. Activity for single-agent alpha-interferon (a-IFN) has been reported in advanced SCCHN4-6 and bio- chemical modulation of 5-FU and cisplatin by a-IFN has also been reported. Preclinical and clinical data in advanced colon cancer suggest that the combination of 5-FU and a-IFN has a synergistic effect, especially in previously untreated patients, with response rates as high as 81‘?0.~-” Similarly, higher than expected re- sponse rates were observed for the combination of 5- FU and a-IFN in chemotherapy-refractory urothelial cancers.” Cisplatin activity also appears to be potenti- ated by a-IFN.I3 Specifically, in nonsmall cell lung can- cer, squamous cell in particular, a response rate of 46% has been rep~rted.’~ These encouraging results stimu- lated a great deal of interest, both in the United States and Europe, in evaluating the potential biomodulation by a-IFN of what is considered the best available com- bination chemotherapy in this disease-5-FU plus cis- platin (given in standard dose and schedule)-in an effort to augment the complete response rate. With this objective, the Southwest Oncology Group designed a Phase I1 trial to evaluate this combination in patients with advanced SCCHN.

Materials and Methods

Patients

The major objective of this study (SWOG 9062) was to determine the complete response rate and toxicity of the combination of 5-FU plus cisplatin plus a-IFN in pa- tients with locally recurrent or metastatic SCCHN. Eli- gible patients had to have histologically proven SCCHN that had persisted or recurred after definitive surgery and/or radiation therapy or have systemic me- tastases. Other requirements for eligibility were no prior chemotherapy, measurable disease, SWOG perfor- mance status of 0-2, and adequate organ function (nor- mal serum creatinine, creatinine clearance of 60 ml/ minute or greater, granulocyte count 1500/ul or greater, platelet count 75,0OO/ul or greater, and ade- quate liver function). All patients gave informed con- sent in accordance with institutional guidelines.

Treatment Plan

All patients were treated with 5-FU 1000 mg/m2/day by continuous intravenous infusion, days 1-4, cisplatin 100 mg/m2 intravenously, day 1, and a-IFN (Roferon- A) at 5 X lo6 units SQ per day, days 1-4. Treatment was repeated every 21 days. The conventional doses of 5- FU and cisplatin were intentionally used so as to maxi- mize the activity of the standard combination and be- cause dose reduction might only result in an activity

equivalent to the standard dose activity despite the syn- ergistic effect of a-IFN. Although we recognized that the toxicity might be increased, knowing the toxicity of the standard combination, we believed that the increase in toxicity would be tolerable and manageable. The a- IFN schedule was designed to maximize exposure to the agent while chemotherapy was being administered and to ensure compliance and convenience. Furthermore, there was no clearly superior schedule for this drug while given with chemotherapy for patients with head and neck cancer. All patients received appropriate hy- dration and mannitol diuresis. 5-Fluorouracil and cis- platin doses were individually modified based on the degree of nadir counts of the previous cycle, stomatitis, dermatitis, ototoxicity, neurotoxicity, and nephrotoxic- ity. Alpha-interferon doses were modified for Grade 2 or higher central nervous system, pulmonary, cardiac, and hepatic toxicities and for grade 3 or higher fever, flu-like symptoms, or hypotension. Recurring toxicity despite dose reduction required the discontinuation of a-IFN. In the absence of progression or significant tox- icity, a minimum of three courses were to be given be- fore evaluation.

Patients achieving complete response were re- quired to receive three additional courses of therapy. Patients with partial response or stable disease were to be treated until disease progression or as determined by toxicity and at the discretion of the managing physician.

Response and Survival Definition Response assessments were based on clinical and radio- logic evaluations. Disease assessment was performed after the first three courses and every two courses there- after. Southwest Oncology Group standard solid tumor response criteriaI5 were used. A complete response (CR) was defined as the dsappearance of all clinical evidence of tumor for a minimum of 4 weeks. Partial response (PR) was defined as 50% or greater decrease in the sum of the products of the two greatest dimensions of mea- sured lesions for a minimum of 4 weeks without the appearance of any new lesions. Stable disease (SD) was defined as a response that does not qualify for CR, PR, or progression. Progression was defined as an increase of 50% over the smallest sum observed in the sum of the products of the two greatest dimensions of measurable lesions, the appearance of new lesions or the reappear- ance of lesions that had disappeared, or a patient’s fail- ure to return for evaluation due to deteriorating condi- tion. Survival was measured from date of registration to date of death.

Statistical Methods This study was designed as a standard Southwest On- cology Group two-stage Phase I1 trial, designed to de-

5-FU, Cisplatin, a-IFN in Advanced SCCHN/Hussain et al. 1235

tect a complete response rate of at least 20% with a power of 92%.

Results

Between August, 1991, and October, 1993, 52 patients were registered onto the study. Two patients were deemed ineligible because of prior colon cancer and lack of documentation of baseline disease (1 patient each). The median age was 59 years (range, 26-77 years), 72% of the patients had a performance status of 0-1, SO% were males, 82% white, 88% had prior sur- gery, and 92% had prior radiation therapy.

One hundred fifty-seven courses of chemotherapy were administered, with a median and a mean of 3 courses (range, 1-12 courses). Twenty-four percent of patients (12/50) received only 1 course of chemother- apy. For these patients, therapy was discontinued due to severe toxicity (3), disease progression (4), and early death unrelated to therapy (5). Twenty-two percent re- ceived two courses; therapy was discontinued due to progression (6) , ototoxicity (2), death due peptic ulcer bleed (l), septic shock (l), and coma (1). Twenty-four percent of patients received 3 courses, 10% received 4 courses, and 20% received 5 or more courses.

Toxicity

All 50 eligible patients were evaluable for toxicity. Two patients died from sepsis complicating Grade 3 or higher granulocytopenia. One of the two patients did not receive a dose reduction for Grade 4 granulocyto- penia at the next scheduled treatment. Thirty-seven pa- tients had Grade 3 or 4 toxicity. Of the 17 patients with Grade 4 toxicity, 12 had hematologic toxicity, 3 stoma- titis, and 2 vomiting. Table 1 illustrates types and grades of therapy-related toxicities.

Response and Survival

All 50 eligible patients were included in the analysis of response and survival. The complete response rate was 6% (3/50; 95% confidence interval, 1%-17%). In addi- tion, 6 (10%) patients achieved a partial response. Three additional patients had unconfirmed responses (1 com- plete and 2 partial) by SWOG standard response cri- teria. These were considered unconfirmed due to lack of a confirmatory reevaluation 4 weeks after the documen- tation of the initial response as required by the SWOG guidelines for response. Ten (20%) patients had stable disease. Thirty-seven of the 50 analyzable patients have died. The estimated median survival is 5 months, with a 95% confidence interval of 4-7 months.

Table 1. Toxicities (N = 50)

Grade (no. of uatients)

Toxicities 1-2 3-4

Hema tologic Anemia 16 4 Granulocytopenia 6 11 Leukopenia 12 22 Thrombocytopenia 6 6

Diarrhea 18 2 Nausea 30 11

Stomatitis 18 10

Fever without infetion 14 1 Creatinine increase 9 1

Nonhematologic

Vomiting 32 3

Malaise/fatigue/lethargy 18 5

Hypokalemia 2 0 Hypomagnesemia 3 1 H y ponatremia 1 1 Ototoxicity 3 4 Peripheral neuropathy 3 1

Discussion

In recent years, biomodulation of chemotherapy by bi- ologic response modifiers has gained major interest and provided an intriguing approach to therapy for patients with advanced cancers. In squamous cell carcinoma of the head and neck, the combination of 5-FU and cis- platin has been used widely. High response rates have been reported in the neoadjuvant setting; however, no increase in the survival rate has been documented when this combination was used neoadjuvantly, adjuvantly, or in recurrent disease, suggesting the need for more active therapy with higher complete response rates. Due to the exciting reports of potential enhancement of 5-FU and cisplatin activity by a-IFN in a variety of chemotherapy nonresponsive tumors, a great deal of enthusiasm has been generated for testing this combi- nation in recurrent/metastatic cancers of the head and neck. Additional support for this combination stems from studies reported by Vokes et a1.16 on escalating dosages of a-IFN added to the combination of 5-FU, cisplatin, and high dose leucovorin (PFL). In their Phase 1/11 study, a 56% complete response rate was reported in 34 patients with previously untreated locally ad- vanced squamous cell carcinoma of the head and neck. This response appeared higher than that observed with two cycles of induction PFL in a prior study conducted by the same investigator^.'^ In the current study, the complete response rate does not appear to be better than that observed with the 5-FW plus cisplatin arm of

1236 CANCER October 2,1995, Volume 76, No. 7

SWOG 8514 (given with same dose and schedule as in the current study) with a comparable median survival, 5 versus 6.6 months, re~pectively.~ Moreover, the overall and the complete response rates in this study are con- sistent with the reported response rates for this combi- nation in patients with recurrent squamous cell cancer of the head and neck, suggesting that a-IFN does not enhance the activity of this combination. Of signifi- cance is that 74% of the patients in the current study experienced at least one Grade 3 or 4 toxicity, compared with only 33% in the 5-FU plus cisplatin arm of SWOG 85 14. Although these differences are not directly com- parable and may be attributable to differences in the conduct and patient populations of the two studies, they are nevertheless important in that they suggest that a-IFN may contribute to increased Grade 3-4 tox- icity. In addition to conventional toxicities, in this study, 10% of patients developed Grade 3 fatigue/malaise/ lethargy. Similar observations of increased high grade toxicity using somewhat different treatment schedules were reported by other investigator^.'^,'^ Although the median number of courses administered in the current study is lower than the median reported for 5-FU plus cisplatin arm of SWOG 8514,3 versus 4, respectively, it is unlikely that this contributed to the lack of apparent enhancement of activity and although 12 (24%) of pa- tients received only l course of therapy, toxicity was a factor in only 3 patients and the remainder either died or obviously progressed.

Benasso et a1.,20 using a 5-day schedule of cisplatin 20 mg/m2/day, 5-FU 200 mg/m2/day intravenous bo- lus, and a-IFN 3 X lo6 units/day IM (a-IFN was also given 3 times/week during the 2 weeks interval among cycles), reported a 54% overall response rate with a 31% complete response rate in 14 previously treated pa- tients. Bensmaine et al.'l observed 2 complete and 4 partial responses in 18 evaluable patients. In contrast, results comparable to ours were reported by Shirinian et al. and The Hoosier Oncology The direct comparison of the outcomes from- studies using the combination of a-IFN, 5-FU, and cisplatin in patients with recurrent SCCHN is made difficult owing to differences in the dose and schedule of all three drugs used in these trials. However, the conflicting response rates observed with this combination may stem from the early reporting of ongoing trials with small sample sizes and the likely differences and heterogeneity of the patient population. Although the schedule of a-IFN may be a factor, it is unlikely to be the cause of the major discrepancy in the reported response rates, because high response rates were reported not only with sched- ules using a-IFN during 5-FU plus cisplatin followed by maintenance a-IFN but also when a-IFN was used only during chemotherapy with no maintenance between

courses.16,21 It is difficult to speculate on the role of leu- covorin in explaining the difference in the response rates between our study and the one reported by Vokes et al. l6 because their study was conducted in previously untreated patients.

The combination of a-IFN, cisplatin and 5-FU, as given in this study, appears to have increased toxicity with no apparent enhancement of response rates asso- ciated with standard dose and schedule of 5-FU and cis- platin. The merits of biomodulation of the 5-FU and cis- platin combination by a-IFN will require the investiga- tion of alternative dose levels and schedules.

References

1.

2.

3.

4.

5.

6.

7.

8.

9.

10.

11.

12.

Al-Sarraf M. Chemotherapeutic management of head and neck cancer. Cancer Metastasis Rev 1987;6:191-8. Schantz SP, Harrison LB, Hong WK. Cancer of the head and neck. In: DeVita VT Jr., Hellman S, Rosenberg SA, editors. Prin- ciples and Practice of Oncology. 4th ed. Philadelphia: JB Lippin- cott Company, 1993;574-630. Forastiere AA, Metch B, Schuller DE, Ensley JF, Hutchins LF, Triozzi P, et al. Randomized comparison of cisplatin plus fluo- rouracil and carboplatin plus fluorouracil versus methotrexate in advanced squamous cell carcinoma of the head and neck: a Southwest Oncology Group Study. ] CIin Oncol 1992; 10:1245- 51. Medenica RN, Slack M. Clinical results of leukocyte interferon- induced tumor regression in resistant human metastatic cancer resistant to chemotherapy and/or radiotherapy-pulse therapy schedule. Cancer Drug Deliv 2:53-76, 1985. Connors JM, Andiman WA, Howarth CB, Liu E, Merigan TC, Savage ME, et al. Treatment of nasopharyngeal carcinoma with human leukocyte interferon. ] C h i Oncol 1985;3:813-7. Vlock DR, Jonson J, Myers E, Day R, Gooding We, Whiteside T, et al. Preliminary trial of non-recombinant interferon alpha in recurrent squamous cell carcinoma of the head and neck. Head Neck 1991; 13:15-21. Wadler S and Schwartz EL. Antineoplastic activity of the combi- nation of interferon and cytotoxic agents against experimental and human malignancies: a review. Cancer Res 1990;50:3473- 86. Chu E, Zinn S, Boarman D, Allegra CJ. Interaction of interferon and 5-fluorouracil in H630 human colon carcinoma cell line. Cancer Res 1990;50:5834-40. Kemeny N, Younes A, Seiter K, Kelson D, Sammarco I', Adams L, et al. Interferon alpha-2a and 5-fluorouracil for advanced co- lorectal carcinoma: assessment of activity and toxicity. Cancer

Pazdur R, Ajani JA, Patt YZ, Winn R, Jackson D, Shepard B, et al. Phase I1 study of fluorouracil and recombinant interferon alfa- 2a in previously untreated advanced colorectal carcinoma. J CIin

Wadler S, Schwartz EL, Goldman M, Lyver A, Rader M, Zim- merman M, et al. Fluorouracil and recombinant alfa-2a-inter- feron: an active regimen against advanced colorectal carcinoma. JCIin Oncol 1989;7:1769-75. Logothetis CJ, Hossan E, Sella A, Dexeus FH, Amato RJ. Fluoro- uracil and recombinant human interferon alfa-2a in the treat- ment of metastatic chemotherapy-refractory urothelial tumors. ] Natl Cancer Znst 1991;83:286-8.

1990;66:2470-5.

Oncd 1990; 8~2027-31.

5-FU, Cisplatin, a-IFN in Advanced SCCHNIHussain et al. 1237

13.

14.

15.

16.

17.

Carmichael J, Fergusson RJ, Wolf CR, Balkwill FR, Smyth JF. Augmentation of cytotoxicity of chemotherapy by human al- pha-interferons in human non-small cell lung cancer xenografts. Cancer Res 1986;46:4916-20. Bowman A, Fergusson RJ, Allan SG, Stewart ME, Gregor A, Cornbleet MA, et al. Potentiation of cisplatin by alpha-inter- feron in advanced non-small cell lung cancer (NSCLC). Ann On-

Green S, Weiss G. Southwest Oncology Group standard re- sponse criteria, end point definitions and toxicity criteria. Invest New Drugs 1992; 10:239-53. Vokes EE, Ratain MJ, Mick R, McEvilly JM, Haraf D, Kozloff M, et al. Cisplatin, fluorouracil, and leucovorin augmented by inter- feron alfa-2b in head and neck cancer: a clinical and pharmaco- logic analysis. ] Clin Oncol 1993; 11:360-8. Vokes EE, Schilsky RL, Weichselbaum RR, Kozloff MF, Panje WR. Induction chemotherapy with cisplatin, fluorouracil, and high-dose leucovorin for locally advanced head and neck can-

C O ~ 1990; 1:351-3.

18.

19.

20.

21.

cer: a clinical and pharmacologic analysis. ] Clin Uncol 1990;8: 241-7. Shirinian M, Choksi AJ, Dimery I, Heyne K, Lippman S, Guillory C, et al. Phase 1-11 study of cisplatin + 5-FU + alpha-interferon for recurrent squamous cell carcinoma of the head and neck [ab- stract]. Proc Am Soc Clin Uncol 1992; 1:249. Arquette MA, Mortimer JE, Loehrer RA. A phase I1 Hoosier On- cology Group trial of interferon alpha-2b added to cisplatin and 5-fluorouracil in recurrent or metastatic head and neck cancer [abstract]. Pror Am SOC Clin Oncol 1994; 13:901. Benasso M, Merlano M, Blengio F, Cavallari M, Rosso R, Toma S. Concomitant alpha-interferon and chemotherapy in advanced squamous cell carcinoma of the head and neck. Am J Clin Oncol 1993; 16(6):465-8. Bensmaine A, Azli N, Domenge C, Armand JP, Vokes EE, Cvit- kovic E. CDDP-5FU modulation by alpha IFN 2 8 (IFN) in meta- static and/or recurrent head and neck squamous cell carcinoma (HNSCC) [abstract]. Ann Oncol 1992;3(5):153.