Neoadjuvant Therapy With Cisplatin/Fluorouracil vs Cisplatin/UFT in Locally Advanced Squamous Cell Head and Neck Cancer

ABSTRACT: This study compared the activity and toxicity of fluorouracil (5-FU)/cisplatin with the combination tegafur and uracil (UFT)/cisplatin in the neoadjuvant treatment of locally advanced-stage III or IV (M0)-head and neck cancer. A total of 67 patients were randomly assigned to treatment with cisplatin 100 mg/m2 on day 1 followed by either a continuous infusion of 5-FU l,000 mg/m2/day on days 2 through 6 (group 1) or oral administration of UFT 300 mg/m2/day on days 2 through 20 (group 2). Both treatments were repeated every 21 days for four cycles. Responding patients received locoregional standard radiotherapy (50 to 70 Gy) after chemotherapy. Group 1 was comprised of 34 patients, 30 of whom were men, with a median age of 57.5 years; 79% of this group had a Karnofsky performance status of 90% to 100%; 70% had a squamous and 29% an undifferentiated histology. The majority (85%) had stage IV disease. Of the 33 patients in group 2, 29 were men. The median age was 56 years. Most (88%) had a performance status of 90% to 100%. More patients had a squamous than an undifferentiated histology (82% vs 18%) and most (88%) had stage IV disease. Overall response in group 1 was 73% (21% complete) compared with 79% (18% complete) in group 2. At a median follow-up of 84 months, no significant differences have emerged in overall survival, 15 vs 37 months, or time to progression, 8.5 vs 14.5 months, for groups 1 and 2, respectively. Toxicity was also similar, except for phlebitis, which occurred significantly more often in group 1 (71% vs 9%). Cisplatin/UFT was as effective as the classic cisplatin/5-FU regimen and has the advantages of outpatient oral administration and a lower incidence of phlebitis.[ONCOLOGY 11(Suppl 10):90-97, 1997]

Introduction

Since the early 1980s, induction or neoadjuvant chemotherapy has been promoted as a new therapeutic approach to the treatment of cancer.[1] Several neoadjuvant chemotherapy trials have been conducted to reduce initial tumor burden, destroy possible pre-existing micrometastases before surgery, and, in some cases, improve overall survival and/or disease-free survival. Most of these studies have involved patients with head and neck squamous cell carcinoma, transitional cell bladder carcinoma, bone and soft-tissue sarcomas, and lately, stage IIIA non-small cell lung cancer (NSCLC).[2-4]

For patients with head and neck cancer, this approach to therapy has the additional potential to allow preservation of organ anatomy and function, as the use of radiotherapy following neoadjuvant chemotherapy could substitute for radical surgery. This treatment would thus afford responding patients a superior quality of life and the possibility of maintaining their normal work and social activities.

The combination of cisplatin (Platinol) followed by fluorouracil (5-FU) as a 120-hour continuous infusion has been considered standard neoadjuvant chemotherapy for head and neck cancer ever since this therapeutic regimen was published by Al-Sarraf.[5] Other investigators have tried to improve on this therapeutic approach using changes in drug combinations or administration methods, with the aim of reducing hospitalization time. The impact on survival of these modifications is not yet clear, however, and their ultimate value in neoadjuvant chemotherapy is still controversial.[6]

The combination of tegafur and uracil (UFT) is an oral formulation that is rapidly absorbed, and has antitumor activity similar to that of 5-FU. Continuous administration of oral UFT has proved to be therapeutically equivalent to 5-FU given by continuous infusion, but it does not require hospitalization.[7] Based on these factors, we have undertaken a prospective study to compare the standard Al-Sarraf regimen with neoadjuvant treatment using cisplatin plus UFT in patients with locally advanced head and neck cancer. Efficacy, toxicity, survival, and time to progression have been analyzed for the two therapeutic approaches.

Patients and Methods

The primary study objective was to compare partial and complete response rates after chemotherapy and after treatment completion and to investigate differences in time to progression and overall survival between the two therapeutic approaches. A secondary goal of the study was to evaluate the toxicity of the two chemotherapy regimens.

Patient Selection

A total of 67 patients diagnosed with locally advanced head and neck carcinoma (squamous cell or undifferentiated) between 1988 and 1992 were included in a randomized prospective study of neoadjuvant chemotherapy. Patients were selected according to the following criteria: confirmed tumor histology, locoregional spread, stage III or IV (M0), Karnofsky index greater than 70%, no previous treatment of any kind, evaluable or measurable tumor lesions, adequate renal and liver functions, no previous neoplasia, and patients’ written consent.

Patients included 59 men and 8 women with a median age of 56.5 years (range, 20 to 75 years) and a median Karnofsky index ranging from 90% to 100%. A weight-loss greater than 5 kg was observed at diagnosis in 26 of 67 patients. Histologic examination confirmed squamous carcinoma in 51 patients and undifferentiated carcinoma in 16. Tumor differentiation was grade 1 or 2 in 31 cases, grade 3 in 17, and unknown in 19. The primary tumor site was the oropharynx in 26 cases, the cavum in 16, the hypopharynx in eight, the larynx in six, and the oral cavity in eight; three patients had cervical lymph node metastases at diagnosis, with unknown primary site. Head and neck carcinoma was stage IV (M0) in 58 patients and stage III in the other nine.

Patients were randomly assigned to treatment according to the classic Al-Sarraf regimen (34 patients) or the cisplatin/UFT regimen (33 patients). Characteristics of the two groups of patients detailed in Table 1, indicated a good balance between them.

Study Design and Therapeutic Schemes

Once selection criteria were confirmed and the patients randomized, the two treatment groups began therapy: 34 patients received cisplatin (100 mg/m²) on day 1, followed by a continuous infusion of 5-FU 1000 mg/m²/day on days 2 through 6; 33 patients received cisplatin 100 mg/m² on day 1, followed by oral administration of UFT at a dose of 300 mg/m²/day in three doses on days 2 through 20. Both treatment regimens were started every 21 days and repeated four times. Responding patients received locoregional standard radiotherapy (50 to 70 Gy) after chemotherapy.

Evaluation of Response

Response to treatment was evaluated after chemotherapy and, in responding patients, again after radiotherapy. Patients underwent a physical examination, blood test, renal and liver function evaluation, chest x-ray, head and neck computed tomography, and endoscopic evaluation of the otorhinolaryngeal area. Complete response (CR) was defined as no evidence of disease for at least 4 weeks, and partial response (PR) as a greater than 50% reduction in measurable tumor mass for the same time period. Stable disease was assumed if tumor reduction was less than 50% or an increase was less than 25%. Progression was inferred in patients with a greater than 25% increase in measurable tumor mass and/or new lesions.

Statistical Analysis

Overall survival was calculated from the beginning of chemotherapy until death, and disease-free survival from the beginning of treatment until the date of relapse or disease-free death. Duration of CR was measured from the date of response until the date of relapse. Partial response was measured from the beginning of treatment until relapse. The Kaplan-Meier method was used to construct overall and disease-free survival curves, and the Mantel-Haenszel test was used to compare response rates, overall survival, and disease-free survival. A value of P < .05 was considered statistically significant.[8]

Results

Responses after four cycles of chemotherapy are summarized in Table 2. In the cisplatin/5-FU arm, the overall response rate was 73.5% (25 of 34 patients), with a CR rate of 20.6%, a PR rate of 52.9%, two patients with stable disease (5.9%) and seven patients with progressive disease (20.5%). In the UFT/cisplatin arm, the response rate was 78.8% (26 of 33 patients), with a CR rate of 18.2%, a PR rate of 60.6%, one patient with stable disease, and six with progressive disease. In both groups, the median number of cycles required for maximum response was three. There were two toxic deaths in each arm, categorized as progressive disease. No statistically significant difference in response was observed between the two arms.

Response to chemotherapy of the primary tumor and involved regional lymph nodes was analyzed separately and the results are presented in Table 3. In the cisplatin/5-FU arm, the primary tumor response rate was 71% (22 of 31 patients), with 29% complete response and 41.9% partial response. Two patients had stable disease (6.5%) and seven had progressive disease (22.6%). In the cisplatin/UFT arm, the tumor response was 84.4% (21.9% complete response and 62.5% partial response), one stable disease (3.1%), and four progressive disease (12.5%). When the data for the two groups were compared, no statistically significant differences were found (P > .05).

When nodal response was analyzed in the cisplatin/5-FU group, the overall response was 76.7% (23/30 cases), with 30% CR and 46.7% PR, two patients with stable disease (6.7%) and five with progressive disease (16.7%). The cisplatin/UFT group had a 76% overall response (19/25 cases), with 28% CR and 48% PR, two with stable disease (8%), and four with progressive disease (16%). Again, no statistically significant differences were found between the two arms (P > .05) (Table 3).

Toxicity of Chemotherapy

Hematologic toxicity was not severe in either group (Table 4). In the cisplatin/5-FU arm, only three cases (8.8%) of grade 4 neutropenia and no cases of grade 3 neutropenia occurred. In the cisplatin/UFT arm, three cases (9%) of grade 4 neutropenia also occurred as did two (6.1%) of grade 3 neutropenia.

Gastrointestinal toxicity, including mucositis, also was mild (Table 5). Emesis reached grade 3 in eight of 34 patients (23.5%) in the cisplatin/5-FU arm, and in three of 33 (9.1%) in the cisplatin/UFT arm. Only one case of grade 4 mucositis occurred, in the cisplatin/5-FU arm (Table 5).

As shown in Table 6, phlebitis developed in 24 of 34 patients (70.6%) in the cisplatin/5-FU arm but in only three of 33 patients (9%) in the cisplatin/UFT arm; these differences were statistically significant (P < .05). Other reported toxicities, including alopecia, ototoxicity, and peripheral neuropathy, were always mild.

Four toxic deaths occurred, two in each arm. In the cisplatin/5-FU arm, one patient died of neutropenic sepsis and the other of acute renal failure. In the cisplatin/UFT arm, one patient died of neutropenic enteritis and the other of a non-neutropenic sepsis followed by acute renal failure.

Response to Chemotherapy and Radiotherapy

In the cisplatin/5-FU arm, nine patients did not receive follow-up radiotherapy, seven due to disease progression and two to toxic death. Among the 25 patients who underwent radiotherapy, 22 responses (17 CRs, and 5 PRs) were observed, disease stabilized in one patient and progressed in two (Table 7). In the cisplatin/UFT arm, six patients did not receive radiotherapy, three due to disease progression, two to toxic death, and one to patient rejection of the radiotherapy. In this arm, 27 patients received radiotherapy, leading to 23 responses (16 CRs and 7 PRs) and four disease progressions (Table 7). When these results were analyzed, no statistically significant differences were found (P > .05).

Toxicity of Radiotherapy

The toxicity associated with the radiotherapy is summarized in Table 8. As expected, toxicity was independent of the previous chemotherapy treatment, and the most frequently encountered toxicities were mucositis and radiodermatitis.

Survival and Time to Progression

Results concerning survival and time to progression are presented in Figure 1. With a median follow-up of 84 months (range, 48 to 120 months), no statistically significant differences in overall survival emerged between the two groups (P > .05). Median survival was 15 months (range, 1 to 77+ months) for patients treated with cisplatin/5-FU and 37 months (range, 1 to 69+ months) for those given cisplatin/UFT. There also were no statistically significant differences in time to progression between the two groups. The median times to progression were 8.5 months (range, 0 to 77+ months) and 14.5 months (range, 0 to 69+ months) for the cisplatin/5-FU arm and for the cisplatin/UFT arm, respectively.

Both overall survival and time to progression have been analyzed separately for responding vs nonresponding patients in each group, first for response to chemotherapy, and then for response to chemotherapy plus radiotherapy. As shown in Figure 2, patients who responded to cisplatin/5-FU had a median overall survival of 38 months (range, 1 to 77+ months) and a median time to progression of 14 months (range, 1 to 77+ months); nonresponding patients had a median survival of 6 months (range, 2 to 12 months) and a median time to progression of 2 months (range, 0 to 8 months). These differences were statistically significant (P < .001). In the cisplatin/UFT group (Figure 3), the median overall survival was 39 months (range, 1 to 69+ months) among responding patients and 9 months (range, 1 to 29 months) among nonresponding patients (P < .05). Median time to progression was 38 months (range, 2 to 69+ months) and 3 months (range, 0 to 18 months) for responding and nonresponding patients, respectively (P < .001).

Overall survival and time to progression in response to the complete treatment (chemotherapy plus radiotherapy) are also shown in Figure 2 and Figure 3. Again, statistically significant differences were found between the two categories. In the cisplatin/5-FU arm (Figure 2), patients who had a complete or partial response after chemoradiotherapy had a median overall survival of 38 months (range, 5 to 77+ months) and non-responders had a median overall survival of 6 months (range, 1 to 12 months) (P < .001). In this group, median time to progression was 25 months (range, 2 to 77+ months) for responders and 1 month (range, 1 to 8 months) for nonresponders, a difference that also was significant (P < .001). In the cisplatin/UFT arm (Figure 3), median overall survival and median time to progression, respectively, were 43 months (range, 4 to 69+ months) and 27 months (range, 3 to 69+ months) for responders, and 7.5 months (range, 1 to 29 months) and 3 months (range, 1 to 18 months) for nonresponders. These differences also were statistically significant (P < .001).

Current Patient Data

With a median follow-up of 84 months (range, 48 to 120 months), 10 patients (29.4%) in the cisplatin/5-FU group are alive without disease, one patient (2.9%) is alive with disease, 20 patients (58.8%) had tumor-related deaths, two patients (5.8%) had toxic or treatment-related deaths, and one patient (2.9%) was lost to follow-up. In the cisplatin/UFT group, nine patients (27.3%) are alive without disease, five patients (15.1%) are alive with disease, 15 patients (45.6%) died due to disease progression, one patient (3%) died of a second tumor, two patients (6%) had toxic deaths, and one patient (3%) was lost to follow-up.

Conclusions

Although the sample size of patients included in this study was small (N = 67), it can be concluded that combination cisplatin/UFT as neoadjuvant chemotherapy for locally advanced head and neck cancer produces a response rate similar to that of the classic cisplatin/5-FU combination. No statistically significant differences in overall survival or time to progression emerged when the two therapeutic schemes were compared. Although significant differences in both overall survival and time to progression were found between responding and nonresponding patients, whether those differences were due to the treatment received is not certain. The variability could be attributed to a patient selection bias related to the natural history of the tumor. Only a large randomized study of neoadjuvant chemotherapy vs standard surgery, with or without radiotherapy, could answer that question.

The tolerability of chemotherapy was similar in both groups, except for the development of phlebitis, which was significantly more common in the cisplatin/5-FU arm.

It can be concluded that the cisplatin/UFT scheme is as effective as cisplatin/5-FU for the neoadjuvant treatment of locally advanced head and neck cancer. The study regimen also has the advantages of reducing certain toxicities, such as phlebitis, and of avoiding the requirement for hospitalization, thus reducing treatment costs and improving the quality of life for patients.

References:

1. Holland JF: Induction chemotherapy: An old term for an old concept, in Neoadjuvant chemotherapy. Colloquium INSERM—John Libbey series. Proceedings 1st International Congress on Neo-Adjuvant Chemotherapy. 137:45-47, 1986.

2. Jacobs C: Adjuvant and neoadjuvant treatment of head and neck cancers. Semin Oncol 18:504-514, 1991.

3. Rosell R, Gómez-Codina J, Camps C, et al: A randomized trial comparing preoperative chemotherapy plus surgery with surgery alone in patients with non-small cell lung cancer. N Engl J Med 330:153-159, 1994.

4. Simon SD, Srougi M: Neoadjuvant M-VAC chemotherapy and partial cystectomy for treatment of locally invasive transitional cell carcinoma of the bladder. Prog Clin Biol Res 353:169-174, 1990.

5. Al-Sarraf M: Chemotherapeutic management of head and neck cancer. Cancer Metastasis Rev 6:181-198, 1987.

6. Schantz SP, Harrison LB, Hong WK: Cancer of the head and neck, in: DeVita VT Jr, Hellman S, Rosenberg SA(eds): Cancer: Principles and Practice of Oncology, 4th ed. pp 574-663. Philadelphia, J B Lippincott Co, 1993.

7. Malik STA, Talbot D, Clarke PI, et al: Phase II trial of UFT in advanced colorectal and gastric cancer. Br J Cancer 62:1023-1025, 1990.

8. Simon R: Design and conduct of clinical trials, in: DeVita VT Jr, Hellman S, Rosenberg SA: Cancer: Principles and Practice of Oncology, 4th. ed. pp 418-440. Philadelphia, J B Lippincott Co, 1993.