Chemotherapy of head and neck cancer

Annals of Oncology 3 (Suppl. 3): S11-S14, 1992. © 1992 Kluwer Academic Publishers. Printed in the Netherlands.

Original article Chemotherapy of head and neck cancer A. A. Forastiere Johns Hopkins Oncology Center, Baltimore, Maryland, U.S.A.

Summary. Numerous studies have shown cisplatin-based chemotherapy to be effective in the treatment of head and neck cancer. Cisplatin/5-fluorouracil (5-FU) is the most frequently used combination regimen, but neurotoxicity, ototoxicity, and renal toxicity limit repeated dosing (for longterm treatment of responding patients) and dose intensification. In studies to date, the analogue carboplatin appears to have activity similar to cisplatin, the advantage of no significant neurotoxicity, ototoxicity, or renal toxicity, and less emetic potential. Two randomized trials have shown cisplatin/5-FU and carboplatin/5-FU superior in terms of response rate compared to single-agent therapy. Treatment with combinations of carboplatin/methotrexate and carbo-

Introduction

Cisplatin is probably the most widely used agent to treat squamous cell carcinoma of the head and neck. Its efficacy is well established for the palliation of recurrent disease and as an adjunct to surgery and radiation therapy in the initial management of newly diagnosed, advanced-stage disease. However, ototoxicity, neurotoxicity, and nephrotoxicity preclude dose intensification and long-term treatment in responding patients. In addition, cisplatin-induced nausea, vomiting, and anorexia often lead to weight loss and a decline in performance status. This is of particular concern in the treatment of patients with head and neck cancer, who as a group often have borderline nutritional and functional assessments. Hence, the more favorable toxicity profile of carboplatin makes it an appealing agent for the treatment of head and neck cancer. Carboplatin is not associated with significant neurotoxicity, ototoxicity, or renal toxicity, has less emetic potential, and can be easily administered in the outpatient setting. For these reasons it has a high rate of patient acceptance. Based on phase I testing and pharmacokinetic parameters, a carboplatin dose of 400 mg/m2 is considered the equivalent in potency to 100 mg/m2 cisplatin. Because of dose-limiting myelosuppression, carboplatin is administered on a 4-week schedule.

Platinum chemotherapy in recurrent disease

The cumulative results of phase II trials evaluating cis-

platin/cisplatin is feasible, but myelosuppression is dose-limiting. As an induction regimen, carboplatin/5-FU yields response rates similar to cisplatin/5-FU. Overall, carboplatin has a more favorable toxicity profile for treating head and neck cancer patients who often have multiple medical problems when presenting for palliation of cancer. Reversible myelosuppression is dose-limiting. However, the availability of hematopoietic growth factors may allow investigators to safely intensify dosing, particularly in combined-modality curative treatment regimens for the newly diagnosed patient.

Key words: head and neck, carboplatin, cisplatin

platin alone or combined with infusional 5-fluorouracil (5-FU) in recurrent disease are shown in Table 1. A higher response rate can be achieved with the combination; this has been confirmed in randomized comparative trials [1, 2]. Table 1. Cisplatin: Phase II trials in recurrent disease. Regimen (mg/m2)

No. evaluable patients

°/o Overall response/% CR

Cisplatin 80-120 q3-4wk

144

33/8

Cisplatin 100 5-FU 1,000/d x 4-5 d q3-4wk

407

50/13

CR = complete response.

Carboplatin as a single agent has been evaluated in 190 patients in phase I and II trials, with an overall response rate of 21%. Doses varied between 200 and 400 mg/m2 administered once every 4 weeks [3-7] and 60 to 80 mg/m2 using a divided-dose schedule (Table 2) [8]. Because not all patients were treated at the maximum tolerable dose, the overall response rate may represent a low estimation. The next logical step in the clinical evaluation of carboplatin was to test it in combination with other cytotoxic drugs with established antitumor activity in head and neck cancer. Two investigators evaluated carboplatin and infusional 5-FU in recurrent disease patients. In a phase I/II trial, Forastiere et al. [9] evalu-

12

Jacobs et al. [2] reported a multicenter trial in which 245 patients were randomized to receive either (1) cisplatin 100 mg/m2 day 1 and 5-FU 1,000 mg/m2 days 1 References Dose (mg/m2) No. evaluable % Overall to 4, (2) single-agent cisplatin 100 mg/m2, or (3) singlepatients response agent 5-FU 1,000 mg/m 2 /d x 4. The overall response rates to the three treatments were 32% for the combi270-500 13 15 [3| 320 14 14 nation and 17% and 13%, respectively, for cisplatin and HI 400 24 29 |5| 5-FU as single-agents. The difference in response rates 200-400 33 27 |6| with the combination vs. the single agents was statisti300-400 70 20 cally significant (P < 0.01). The median durations of 60-80 qd x 5 31 26 [8] response and survival were not different; however, an analysis of patients surviving longer than 9 months reated carboplatin in doses ranging from 240 to 420 mg/ vealed a significant advantage for cisplatin/5-FUm2 in combination with 5-FU 1,000 mg/m 2 /d as a treated patients compared with patients who received 120-hour infusion every 4 weeks. Doses were escalated single-agent cisplatin (P < 0.05). within patients to achieve at least grade 2 myelosupCombinations of carboplatin with either methotrexpression. There were 3 complete responses (CRs) and ate or cisplatin have been evaluated in phase II trials. 10 partial responses (PRs), for a 48% overall response Eisenberger and colleagues [12] randomized 40 parate in 27 patients. The median duration of response tients with recurrent squamous cell carcinoma of the was 4.7 months. As expected, myelosuppression was head and neck to treatment with either weekly methodose limiting. Nonhematologic toxicity was limited to trexate 40 mg/m2 or carboplatin 400 mg/m2 q4wk mild to moderate nausea, vomiting, and stomatitis. plus weekly methotrexate (40 mg/m2). A 25% overall Outpatient treatment was feasible with the use of response rate was observed for both treatment groups, ambulatory infusion pumps, and patient acceptance of but myelosuppression necessitated carboplatin dose repeated courses was excellent. The recommended reductions to 300 and 200 mg/m2. Thus, treatment dose for phase II trials of carboplatin/5-FU in recur- with optimal doses of each agent was not possible due rent disease patients was 300 mg/m2 for poor-risk pa- to overlapping myelotoxicity, and the carboplatin/ tients and 360 mg/m2 for good-risk patients. Olver et methotrexate combination was not recommended for al. [10] conducted a phase II trial of carboplatin/5-FU further study. using 70 mg/m 2 /d x 5 and a 120-hour infusion of Two investigators attempted to combine carboplatin 5-FU every 4 weeks. A 31% response rate was ob- and cisplatin in recurrent disease patients [13, 14]. The served in 51 patients (4 CRs and 10 PRs). The median regimen tested was carboplatin 350 mg/m2 day 1 plus survival for all patients was 4.8 months. Myelosuppres- cisplatin 50 mg/m2 day 1 q4wk. Dimery et al. [13] resion and stomatitis were the major toxicities. ported a 24% overall response rate in 21 patients. The A number of combination chemotherapy regimens median nadir counts were 52 x 109/L for platelets and have been compared with single-agent methotrexate or 1.1 x 109/L for granulocytes. Myelosuppression necescisplatin in prospective randomized trials [11]. Two sitated decreasing the starting dose to 300 mg/m2. trials that evaluated cisplatin/5-FU demonstrated a sig- Powell et al. [14] observed a 33% overall response rate nificant improvement in response rate for the combina- in 22 patients using the same regimen; grade 3 or 4 tion compared with single agents [1, 2]. The Southwest myelosuppression occurred in 7 patients (32%). In Oncology Group randomized 277 patients with recur- these two trials there was no potentiation of nephrorent and metastatic squamous cell carcinoma to one of toxicity, ototoxicity, or neurotoxicity. Myelosuppression three treatments: (1) cisplatin 100 mg/m2 day 1 plus was significant, suggesting that the administration of 5-FU 1,000 mg/m2 days 1 through 4 q3wk; (2) carbo- both drugs on the same day may alter carboplatin renal platin 300 to 360 mg/m2 day 1 plus 5-FU 1,000 mg/ clearance and enhance bone marrow toxicity [15]. Alm2 days 1 through 4 q4wk; or (3) methotrexate 40 to though treatment with both carboplatin and cisplatin is 50 mg/m2 weekly. The primary objective of this trial feasible, no therapeutic benefit in terms of response or was to compare each platinum combination chemother- survival has yet been shown. apy regimen with methotrexate separately. The overall response rates were 32% for the cisplatin treatment, 21% for carboplatin, and 10% for methotrexate. The Platinum chemotherapy induction regimens difference in response rates to cisplatin/5-FU vs. Chemotherapy given prior to surgery or radiation thermethotrexate was significant (P < 0.001), and the differ- apy, termed induction or neoadjuvant treatment, is one ence for carboplatin/5-FU vs. methotrexate was of strategy that has been widely tested to improve survival borderline statistical significance (P = 0.05). Median of newly diagnosed stage III and IV head and neck candurations of response and survival were similar for all cer patients. To date, randomized trials have not shown three treatments. Hematologic and nonhematologic significant differences in survival when induction toxicities were significantly greater with cisplatin/5-FU chemotherapy precedes definitive local therapy [16]. compared with methotrexate; toxicity from carboplatin However, a decreased rate of distant metastases [17— was intermediate compared with the other regimens. 19] and the ability to preserve organ function [19] have Table 2. Phase I and II trials of carboplatin in recurrent head and neck cancer.

m

13

been shown with cisplatin-based chemotherapy. Stud- Table 4. Carboplatin induction regimens.11 ies of cisplatin 80 to 120 mg/m2 combined with infuNo. Regimen (mg/m7d) sional 5-FU 800 to 1,000 mg/m2/d x 5 have shown Referevaluable ences that CR is achieved in approximately 35% of patients patients (range, 19% to 54%) and the overall response rate is Carboplatin 300 85% (range, 73% to 94%) [11, 20-27]. Individual stud- [28] 55 5-FU 1,000 x 5 ies are detailed in Table 3. Table 3. Results of induction chemotherapy with cisplatin/5-FU in advanced head and neck cancer patients. References

No. evaluable patients

Regimen

% Overall response/ %CR

[21]

26

Cisplatin 100 mg/m2 5-FU 1 g/mVd x 5 (2 cycles)

88/19

[22]

88

Cisplatin 100 mg/m2 5-FU 1 g/m 2 /d x 5 (3 cycles)

94/54

[23]

103

Cisplatin 100 mg/m2 5-FU 1 g/m7d x 5 (3 cycles)

87/35

[24]

30

Cisplatin 100 mg/m2 5-FU 1 g/m 2 /d x 5 (3 cycles)

83/43

[25|

42

Cisplatin 100 mg/m 2 5-FU 1 g/m7d x 5 (3 cycles)

86/38

[261

31

Cisplatin 80 mg/m2 5-FU 800 mg/m7dx 5 (2-3

84/23

[271

53

Cisplatin 20 mg/m 2 /d x 6 73/30 5-FU 1 g/m7d x 5 (2-4 cycles)

% Overall response (% CR) 87 (27)

[29]

Carboplatin 360 5-FU 1,000 x 5

22

59(18)

[30|

Carboplatin 250-400 5-FU 1,000 x 5 Carboplatin <330 Carboplatin >330

44

1311

Carboplatin 400 5-FU 1,000 x 5

29

81 (29)

[32]

Carboplatin 300 Tegafur 1,000 x 21

27

77 (33)

24 59

a

Response to chemotherapy prior to surgery and/or radiation therapy. Table 5. Randomized induction platinum chemotherapy trials.

References

Regimen (mg/m 2 )

No. evaluable patients

% Overall response

|33]

Cisplatin 100 + 5-FU 1,000/dx 120h q3wk

18

83

Carboplatin 400 + 5-FU 1,000/dx 120hq3wk

20

67

Cisplatin 100 + 5-FU 1,000/dx 120h q3wk

61

79

From Urba and Forastiere with permission [20]. [34]

At least five phase II trials of carboplatin/5-FU have Carboplatin 400 + 5-FU 58 71 been reported (Table 4) [28-32]. Volling et al. [28] ad1,000/d x 120h q4wk 2 ministered carboplatin 300 mg/m and a 5-day 5-FU infusion for two or three cycles prior to surgery or radiation therapy in 55 patients with advanced head and neck cancer. The overall response rate to chemo- ferences in response rates [33, 34], median survival, therapy was 88%, with 33% of patients achieving CR. and relapse-free survival [34]. A retrospective comparison to 25 patients with similar stage disease treated with cisplatin/5-FU demonstrated Conclusion no difference in response rates but less toxicity and better patient acceptance of the carboplatin regimen. Kish Cisplatin and carboplatin appear similar in efficacy et al. [29] observed a 59% overall response rate to three both as single agents and in combination with 5-FU for cycles of induction carboplatin 360 mg/m2 and 5-FU. the palliation of recurrent disease or as induction therThis was comparable with their experience with cispla- apy. In two randomized trials of recurrent disease patin/5-FU in similar patients with advanced stage IV tients, significantly higher response rates were observed disease. for cisplatin/5-FU compared with single-agent cisplaRosseau and colleagues [30] evaluated carboplatin tin [2], 5-FU [2], or methotrexate [1], and for carbopla250 to 400 mg/m2 prior to surgery or radiation ther- tin/5-FU compared with methotrexate [1]. Data from apy. A 24% response rate was observed in 17 patients recurrent disease trials [8, 12-14] indicate that myeloreceiving <330 mg/m2. In contrast, a 59% response suppression is the limiting factor in combining carborate occurred in 27 patients treated with >330 mg/m2, platin with other cytotoxic drugs, including 5-FU, suggesting a dose-response effect. Preliminary results methotrexate, and cisplatin. from Urba et al. [31] and Machengs et al. [32) are simiIn contrast to the neurotoxicity and ototoxicity that lar, with overall response rates of approximately 80% limit cisplatin administration, hematopoietic growth and CR in about one-third of patients. factors may offer the potential to overcome the myeloThe preliminary results of two randomized trials suppressive effects of carboplatin. This is of particular directly comparing cisplatin/5-FU and carboplatin/ importance in exploiting the dose-response effects sug5-FU induction chemotherapy (Table 5) show no dif- gested by some studies with platinum compounds.

14 Phase I studies are in progress testing carboplatin with interleukin-3 alone and interleukin-3 sequenced with granulocyte-macrophage colony-stimulating factor. This is one experimental approach that, if successful, may result in high CR rates and improvement in survival. References 1. Forastiere AA, Metch B, Schuller DE et al. Randomized comparison of cisplatin plus 5-FU and carboplatin plus 5-FU vs methotrexate in advanced squamous cell carcinoma of the head and neck: a Southwest Oncology Group Study. J Clin Oncol 1992; 10: 1245-51. 2. Jacobs C, Lyman G, Velez-Garcia E et al. A phase III randomized study comparing cisplatin and fluorouracil as single agents and in combination for advanced squamous cell carcinoma of the head and neck. J Clin Oncol 1992; 2: 251-63. 3. Ohnuma T, Leyvraz S, Coffey V et al. Carboplatin: Activity in patients with head and neck, renal cell and ovarian carcinomas (abstr). Proc Am Assoc Cancer Res 1984; 25: 179. 4. Creekmore SP, Micetich KC, Vogelzang N et al. Low toxicity and significant tumor responses in phase II trials of carboplatin in head and neck cancer, non-small cell lung, urothelial, and ovarian cancers. Proc Am Soc Clin Oncol 1985; 4:144. 5. Al-Sarraf M, Metch B, Kish J et al. Platinum analogs in recurrent and advanced head and neck cancer: a Southwest Oncology Group and Wayne State University study. Cancer Treat Rep 1987; 71: 723-6. 6. Calvert AH, Grumbrell LA, Henk MJ. Paraplatin phase II studies in head and neck cancer. Proceedings of the Symposium 'Paraplatin: The British Transition'. London 1986: 25. 7. Inuyama Y, Togawa K, Morita M et al. Phase II study of carboplatin in head and neck cancer. Gan To Kagaku Ryoho 1988; 15:2131-8. 8. Eisenberger M, Hornedo J, Silva H et al. Carboplatin (NSC241-240): an active platinum analog for the treatment of squamous cell carcinoma of the head and neck. J Clin Oncol 1986; 4: 1506-9. 9. Forastiere AA, Natale RB, Takasugi BJ et al. A phase I-1I trial of carboplatin and 5-fluorouracil combination chemotherapy in advanced carcinoma of the head and neck. J Clin Oncol 1987; 5: 190-6. 10. Olver IN, Dalley D, Woods R et al. Carboplatin and continuous infusion 5-fluorouracil for advanced head and neck cancer. Eur J Cancer Clin Oncol 1989; 25:173-6. 11. Al-Sarraf M. Head and neck cancer: chemotherapy concepts. Semin Oncol 1988; 15: 70-85. 12. Eisenberger M, Krasnow S, Ellenberg S et al. A comparison of carboplatin plus methotrexate versus methotrexate alone in patients with recurrent and metastatic head and neck cancer. J Clin Oncol 1989; 7: 1341-5. 13. Dimery IW, Brooks BJ, Winn R et al. Phase II trial of carboplatin plus cisplatin in recurrent and advanced squamous cell carcinoma of the head and neck. J Clin Oncol 1991; 9: 1—6. 14. Powell BL, Stanley V, Brockschmidt J et al. Combination carboplatin and cisplatin for advanced squamous carcinoma of the head and neck (abstr 693). Proc Am Soc Clin Oncol 1990; 9: 179. 15. Trump DL, Brem JL, Tutsch KD et al. Platinum analogue combination chemotherapy: cisplatin and carboplatin - A phase I trial with pharmacokinetic assessment of the effect of cisplatin administration on carboplatin excretion. J Clin Oncol 1989; 5: 1281-9. 16. Forastiere AA. Randomization trials of induction chemotherapy: a critical review. Hematol Oncol Clin North Am 1991; 5 (4): 725-36. 17. Head and Neck Contracts Program. Adjuvant chemotherapy

18.

19.

20. 21.

22. 23. 24. 25. 26. 27. 28. 29.

30.

31.

32.

33.

34.

for advanced head and neck squamous carcinoma. Cancer 1987; 60: 301-11. Schuller DE, Metch BE, Mattox D et al. Preoperative chemotherapy in advanced resectable head and neck cancer: final report of the Southwest Oncology Group. Laryngoscope 1988; 98: 1205-11. The Department of Veterans Affairs Laryngeal Cancer Study Group. Induction chemotherapy plus radiation compared with surgery plus radiation in patients with advanced laryngeal cancer. N Engl J Med 1991; 324: 1685-90. Urba SG, Forastiere AA. Systemic therapy of head and neck cancer: most effective agents, areas of promise. Oncology 1989; 3: 79-88. Kish J, Drelichman A, Jacobs J et al. Clinical trial of cisplatin and 5-FU infusion as initial treatment of advanced squamous cell carcinoma of the head and neck. Cancer Treat Rep 1982; 66:471-4. Weaver A, Flemming S, Kish J et al. Cis-platinum and 5-fluorouracil as induction therapy for advanced head and neck cancer. Am J Surg 1982; 144:445-8. Thyss A, Schneider M, Santini J et al. Induction chemotherapy with cisplatinum and 5-fluorouracil for squamous cell carcinoma of the head and neck. Br J Cancer 1986; 54: 755-60. Jacobs C, Goffinet DR, Goffinet L et al. Chemotherapy as a substitute for surgery in the treatment of advanced resectable head and neck cancer. Cancer 1987; 60: 1178-83. Jacobs JR, Pajak TF, Kinzie J et al. Induction chemotherapy in advanced head and neck cancer. Arch Otolaryngol Head Neck Surg 1987; 113: 193-7. Amrein PC, Weitzman SA. Treatment of squamous cell carcinoma of the head and neck with cisplatin and 5-fluorouracil. J Clin Oncol 1985; 3: 1632-9. Clark JR, Fallon BG, Dreyfuss Al et al. Chemotherapeutic strategies in the multidisciplinary treatment of head and neck cancer. Semin Oncol 1988; 15 (suppl 3): 35-44. Volling P, Schroder M, Rauschning W et al. Carboplatin: the better platinum in head and neck cancer? Arch Otolaryngol Head Neck Surg 1989; 115: 695-8. Kish JA, Ensley JF, Zapazoglow E et al. Carboplatin (CBDCA) (NSC-241240) and 5-FU infusion as an effective outpatient induction chemotherapy in patients with locally advanced head and neck cancer (abstr). Proc Am Soc Clin Oncol 1989; 8: 173. Rosseau F, Hamoir M, Derweduwen K et al. Carboplatin and continuous 5-FU infusion as induction chemotherapy in untreated advanced squamous cell carcinoma of the head and neck. Presented at the Second International Conference on Head & Neck Cancer, Boston, MA, July 31-August 5,1988. Urba S, Spaulding M, Dibb C et al. Induction chemotherapy with carboplatin and 5-fluorouracil in a treatment schedule for organ preservation in patients with advanced resectable head and neck cancer: toxicity and tumor response (abstr). Proc Am Soc Clin Oncol 1991; 10:206. Machengs I, Sanchiz F, Centelles M et al. Head and neck cancer: neoadjuvant chemotherapy with florafur and carboplatin. A preliminary report (abstr). Proc Eur Conf Clin Oncol 1987; 4:950. De Andres L, Lopez Pousa A, Burgues J et al. Randomized trial of neoadjuvant cisplatin and fluorouracil versus carboplatin and fluorouracil in patients with advanced head and neck cancer (abstr). Proc Eur Conf Clin Oncol 1987; 4: 956. Senor CM, Manzano H, Guma J et al. Comparison of cisplatinum and carboplatin and continuous 5-fluorouracil infusion in advanced head and neck cancer (abstr). Proc Eur Conf Clin Oncol 1989;5:P-0358.

Correspondence to: Arlene A. Forastiere, MD Johns Hopkins Oncology Center 600 N. Wolfe St Baltimore, MD 21205