Gemcitabine/Cisplatin Shows Good Response Rate And Favorable Toxicity Profile in Advanced NSCLC

ASCO LOS ANGELES--A phase II study of gemcitabine, a novel nucleosideanalog, plus cisplatin (Platinol) in patients with stage III orIV non-small-cell lung cancer (NSCLC) showed an encouraging 42%response rate and 8-month median survival.

"The median survival is similar to other active combinations,and, given the favorable toxicity profile of this predominantlyoutpatient regimen, we feel it should be further assessed in arandomized phase III study," said W.P. Steward, MD, of theCanadian National Cancer Institute and Queens University, Kingston,Ontario.

Dr. Steward reported the results of the multicenter study, involvingresearchers in the United Kingdom and France, at a scientificsession of the American Society of Clinical Oncology (ASCO) meeting.

At the same session, Italian researchers presented data from theirphase II trial using the same drug doses in a slightly differentschedule. They achieved an even more impressive response rate(54%), and described their regimen as well tolerated on an outpatientbasis despite a considerable incidence of thrombocytopenia.

Gemcitabine (Gemzar from Eli Lilly and Company) is currently availablein the United States for patients with advanced pancreatic cancerthrough a treatment investigational new drug (IND) program.

Rationale for the Combination

Dr. Steward said that the rationale for combining the two drugsstems from their known activity in the disease (gemcitabine producesresponse rates on the order of 24% when given as a single agent),their largely nonoverlapping toxicity profiles, and encouragingpreclinical data suggesting synergy between the two agents ina variety of tumor models, including adenocarcinoma lines.

Based on their phase I study, the researchers chose a regimenof 1 g/m² of gemcitabine given as a 30-minute IV infusionweekly for 3 weeks. Cisplatin, 100 mg/m², was added as a4-hour IV infusion immediately after the third gemcitabine infusion,followed by a 2-week break before the next cycle. The course wasrepeated every 4 weeks.

Sixty patients with previously untreated, inoperable advancedNSCLC were enrolled in the phase II trial. The patients had predominantlysquamous cell tumors, and nearly all had stage IIIb or IV disease,Dr. Steward said.

Of 52 patients evaluable for response, one had a complete responseand 21 a partial response for an overall response rate of 42%.The median duration of response was just under 11 months. Mediansurvival for the entire group was 8 months, and for the responders,14.3 months. There was no difference in survival between patientswith stage IIIb and stage IV disease, Dr. Steward said.

All 60 patients were evaluated for toxicity. The major hematologictoxicity was neutropenia with about 50% of patients experiencinggrade 3 or 4 neutropenia. "However," Dr. Steward said,"no patient required hospitalization for febrile neutropeniaor infection." The neutropenia was of short duration, 2 to3 days. Uncomplicated grade 4 thrombocytopenia was seen in 7%of patients, and grade 3 in 20%.

Overall, the regimen was well tolerated, he said. Alopecia wasrarely a problem, and nausea and vomiting were well controlledwith antiemetics.

The Italian Lung Cancer Group employed a regimen of gemcitabine,1 g/m² weekly for 3 weeks, followed by a week of rest, withcisplatin, 100 mg/m², given on day 2 of each 28-day cycle.

Dr. Lucio Crinò, of Perugia, reported that of 48 enrolledpatients, all with previously untreated, unresectable stage IIIbor IV NSCLC and good performance status (0-1), one patient hada complete response and 25 had a partial response, for an overallresponse rate of 54%. Again, as in the Canadian study, there wasno difference in the response rate between patients with stageIIIb and IV disease.

Dr. Crinò pointed out that two patients with a partialresponse underwent further treatment (surgery or radiation therapy)to render them free of disease.

Long-Lasting Responses

The median duration of response was 51 weeks, and 10 patientshave a long-lasting response in excess of 10 months, includingthe complete responder, Dr. Crinò said. Median time toprogression was 41 weeks, "and we have not yet reached mediansurvival, with median follow-up of 11 months," he said.

Dr. Crinò noted that 52 doses were reduced and 92 omitted,primarily because of grade 3/4 thrombocytopenia, which occurredin 52% of patients. However, he pointed out, the nadir was short,there were no bleeding episodes, and patients were able to resumetherapy at the next cycle.

Although 36% of the patients in this study had neutropenia, thecondition was transient and not associated with infection or neutropenicfever. Nonhema-tologic toxicity was mild and due mainly to theknown side effects of cisplatin, Dr. Crinò said.

The Italian group plans further evaluation of the regimen, witha randomized trial scheduled to start in the next few months,Dr. Crinò said.