Clinical Trial Result Information
Protocol number
BO16411
Title of Study
A Randomized, Double-Blind, Placebo Controlled, Multicenter, Phase III Study of Tarceva (Erlotinib) plus Chemotherapy (Cisplatin and Gemcitabine) Versus Chemotherapy Alone in Patients with Advanced (Stage IIIB or IV) Non–Small-Cell Lung Cancer (NSCLC) Who Have Not Received Prior Chemotherapy
Sponsor
Hoffman-La Roche AG
Company division
Pharmaceutical
Product name
Tarceva
Generic name
erlotinib
Therapeutic area
Non-Small Cell Lung Cancer
Clinical study summary
This randomized, placebo-controlled, multicenter survival study comprised the recruitment of an initial 50-patient safety cohort followed by independent interim analysis to determine the appropriate dose to investigate before the main part of the study.
Study center(s)
The study was recruited from 164 centers in 27 countries.
Phase of development
III
Objectives
Primary: To evaluate the efficacy and safety of erlotinib (Tarceva) in combination with chemotherapy (cisplatin and gemcitabine) versus chemotherapy alone.
Secondary: Secondary objectives included time to progression, response rates, duration of response, pharmacokinetic (PK) and pharmacodynamic parameters, quality of life (time to symptomatic progression) and correlation of EGFR and other markers with response and survival.
Methodology
After eligibility screening, patients with Stage IIIB or IV NSCLC were randomized to receive either Tarceva or placebo in combination with chemotherapy (cisplatin plus gemcitabine). Patients entered into the initial safety cohort received a 100-mg daily dose of Tarceva during a 7-day run-in period and for the first 15 days of the first chemotherapy cycle, thereafter receiving a 150-mg daily dose. Following confirmation of acceptability by an independent safety monitoring board (DSMB), all subsequent patients started with the 150-mg daily dose and did not undergo a run-in period.
Chemotherapy continued for up to 6 cycles, and Tarceva/placebo was continued until disease progression, unacceptable toxicity, or death. Patients withdrawn from study medication were followed until death. Tumor measurements, for assessment of response rates, were performed at 6-week intervals during chemotherapy, and at 12-week intervals thereafter. ECOG status and the lung cancer symptom scale (LCSS) were completed at each visit. Safety was assessed through the monitoring of adverse event information in all patients, and from clinical laboratory tests, from samples taken at each visit, in the first 400 randomized patients.
Number of patients (planned/analyzed)
1172 randomized
Diagnosis and main criteria for inclusion
Entry criteria included histologically documented, inoperable, unresectable, incurable, locally advanced, recurrent, or metastatic (Stage IIIB or Stage IV) NSCLC; no prior chemotherapy or systemic anti-tumor therapy; Eastern Cooperative Oncology Group (ECOG) performance status of 0–1. Selected exclusion criteria included prior chemotherapy or therapy with systemic anti-tumor therapy, prior exposure to agents directed at EGFR family members, unstable systemic disease, symptomatic or untreated brain metastases, granulocyte count <= 1500/mm3, platelet count <100,000/mm3, serum bilirubin>1.5 x upper limit of normal (ULN), serum ALT and AST >1..5 x ULN (>5 x ULN if due to liver metastases), serum creatinine >1.5 x ULN or creatinine clearance <60mL/min, pregnancy.
Test product, dose and mode of administration or test procedure
In the initial 50 patient safety cohort, patients were randomized to receive Tarceva, (100mg/day) or placebo alone during a 7 day run-in period, following which the treatments were combined with chemotherapy. The dose of Tarceva was escalated to 150mg once daily from day 15 of the first chemotherapy cycle, provided that the initial 100mg dose was tolerated.
During the exploratory stage, when the best dose for the study was being selected, Tarceva was given at 100mg/po/qd/7 days followed by 150mg/po/qd/day 15 onward, until disease progression, toxicity, or death.
Following dose selection for the study (after interim analysis) Tarceva was given at 150mg/po/qd until disease progression, toxicity or death.
Duration of treatment
Tarceva: daily treatment from Day 15 on (150 mg) until disease progression, unacceptable toxicities, or death. Gemcitabine/Cisplatin: 6 cycles.
Reference therapy, dose and mode of administration or reference procedure
Six 21-day cycles of iv chemotherapy (gemcitabine 1,250mg/m2 on days 1 and 8 and cisplatin 80mg/m2 on day 1).
Criteria for evaluation (efficacy, safety)
Primary efficacy parameter: Duration of survival.
Secondary efficacy parameters included time to disease progression or death (TTP), response rate, duration of response, and quality of life (using the Lung cancer Symptom Scale).
PK parameters: Intensive pharmacokinetic monitoring of Tarceva, gemcitabine and cisplatin were conducted on samples from 24 patients in selected centres, 12 from the Tarceva group during the initial safety cohort part of the study, and 12 from the placebo group during the main part of the study.
Safety parameters: Adverse events, clinical laboratory tests (first 400 randomized patients only).
Statistical methods
The primary endpoint was duration of survival in the intent-to-treat (ITT) population, which was analyzed using a 2-sided log rank test. The study was designed to have 80% power for the 2-sided hypothesis that addition of Tarceva to chemotherapy would increase median survival by 25% relative to survival reported for cisplatin and gemcitabine alone (from 9.1 to 11.375 months; hazard ratio 0.8).based on these assumptions, 643 events were required, corresponding to enrollment of 1116 patients.
A 2-sided log-rank test was used to compare TTP and time to symptomatic progression between the two treatment arms, and the objective response rates were compared using a chi-squared test. Kaplan-Meier survival methodology was used to estimate medians and 95% confidence limits for duration of survival, TTP, and duration of response.
Analysis of safety ws performed on all patients who received at least one dose of the trial medication, and had at least one safety follow-up. Pharmacokinetic data were analyzed descriptively, and plasma concentration versus time data were analyzed by non-compartmental methods.
Summary (efficacy, safety, other results)
Efficacy – The primary endpoint of the study (overall survival) was analyzed on the first 643 deaths reported, as defined in the protocol. There was no meaningful difference in overall survival between the 2 treatment groups – 309 days [95% CI: 282, 343] (placebo) versus 301 days [95% CI: 274, 315] (Tarceva), giving a hazard ratio of 1.06. Data from all 690 deaths on the database at cut-off were analogous: 308 days (placebo) versus 298 days (Tarceva), hazard ratio 1.03. Within the limits arising from the variations in numbers of patients, none of the subgroup analyses suggested that any particular subpopulation either benefited or suffered to a greater extent than the overall population, although there is a suggestion that female patients receiving Tarceva plus chemotherapy have a slightly better prognosis than their counterparts receiving placebo (hazard ratio 0.83, 95% CI: 0.58, 1.18).
There was no benefit of Tarceva treatment over placebo treatment on the time to progressive disease (PD) or death in patients while receiving study drug (178 days [placebo] vs 167 days [Tarceva]). The analysis of the female subpopulation for time to PD or death on study drug produced a hazard ratio of 0.63 (95% CI: 0.46, 0.88). Analysis of time to PD or death “overall,” (ie, including the follow-up period) again showed no benefit of Tarceva treatment (172 days [placebo] vs 166 days [Tarceva]). The apparent benefit seen in the subgroup of female patients in the “on study drug” analysis was maintained in this overall analysis, with very similar results: the hazard ratio was 0.69 (95% CI: 0.52, 0.92).
Response rates were measured in patients who had measurable lesions at baseline. The proportion of patients with objective responses (complete response [CR] + partial response [PR]) was essentially the same for each treatment group (31.5% in the Tarceva group and 29.9% in the placebo group), although very few patients exhibited CRs. More than 50% of patients in both groups had either a PR or stable disease, with only small differences between groups for each category. Data from the 376 analyzable samples showed no correlation between HER1/EGF-R expression (graded as 0, 1+, 2+, or 3+) and either survival or response. There was only a small difference in the median duration of response, with a median duration of almost 6 months seen in the 2 treatment groups. The time to symptomatic progression was defined as a worsening from baseline in the average symptom burden index by >=25%. There was essentially no difference in the median time to symptomatic progression seen in Tarceva-treated patients relative to those receiving placebo (68 days vs 76 days).
PK – No evidence was found for a PK interaction between Tarceva, cisplatin, and gemcitabine based on estimated mean exposure ratios and 95% CIs
Safety – Overall, gastrointestinal disorders were the most frequently reported events, and occurred more frequently in the Tarceva group (75% vs 69% of patients in the placebo group). There was a marked elevation in the incidence of diarrhea in the Tarceva group (40% vs 17% in the placebo group). The most notable imbalance between the groups related to the incidence of skin disorders (70% vs 34%), arising mainly from the known association between Tarceva treatment and rash (66% vs 19% in those receiving placebo). Other body systems in which there was a higher incidence of events on active treatment were metabolism and nutritional disorders (38% vs 33%; mainly driven by anorexia and dehydration), investigations (28% vs 21%; mainly driven by weight decrease), renal and urinary disorders (13% vs 7%; mainly renal failure), and eye disorders (9% vs 5%; mainly conjunctivitis). Overall, there was a higher incidence of NCI-CTC (v.2) Grade 3 and 4 events in the Tarceva group (77%) than in the placebo group (72%).
The proportion of deaths, together with the distribution of their cause as due to PD or adverse event, was very similar between treatment groups (death in 48% of patients due to PD in the Tarceva group vs 46% in the placebo group; death in 11% of patients due to adverse events in the Tarceva group vs 12% in the placebo group). Almost all events leading to death were considered by the investigators as unrelated to study medication and were attributed to disease progression (268 in the placebo arm and 277 in the Tarceva arm); however, there was an increased incidence of events considered as probably related to study medication in the Tarceva group. Some events resulted in death only for patients in the Tarceva group. Notable among these are renal failure and acute renal failure, from which 4 patients died; febrile neutropenia, neutropenic sepsis, and neutropenia, from which 4 patients died; and diarrhea, from which 2 patients died.
The incidence of serious adverse events was higher in the Tarceva group (53%) than in the placebo group (47%). This was driven by an excess of serious events in the Tarceva group in 4 body systems: blood and lymphatic system disorders (mainly anemia and thrombocytopenia), gastrointestinal disorders (mainly diarrhea), renal and urinary disorders, and skin and subcutaneous tissue disorders. There was a marked increase in the incidence of serious renal disorders in the Tarceva group relative to the placebo group (17 patients [3%] vs 5 patients [<1%]). Renal failure and acute renal failure were the major events behind this, recorded in 12 Tarceva patients but in only 2 placebo patients. This is likely to have been due to Tarceva-related diarrhea, which, when combined with poor or inadequate hydration of patients on a cisplatin, contributed to an exacerbation of cisplatin toxicities.
There was a higher proportion of patients from the Tarceva group who had adverse events leading to discontinuation of treatment (22% vs 17% in the placebo group), which was also reflected in an increased number of such adverse events (161 vs 128). This difference is mainly accounted for by the number of patients withdrawing as a consequence of skin disorders in the Tarceva group (27/580 [5%]) compared with the number withdrawing in the placebo group (3/579 [<1%]). The overall incidence of withdrawals due to blood and lymphatic system disorders was slightly reduced in the Tarceva group (2 patients vs 7 patients in the placebo group), despite the higher incidence of serious adverse events within this body system in the Tarceva group.
Only two cases of interstitial lung disease (ILD) were recorded as adverse events in this trial; both occurred in patients in the placebo group, and were grde 2 in intensity. One patient in the Tarceva group died from atypical primary pneumonia, which was considered remotely related to treatment. The post-mortem findings of this patient included ILD. Given the difficulty in diagnosing ILD solely on clinical grounds, a review of all adverse events in the respiratory body system was performed, which did not reveal any cause for concern regarding the development of ILD in Tarceva-treated patients in this trial.
For most of the hematology parameters, there were trends to higher numbers of patients in the Tarceva group with significant shifts in NCI-CTC Grade. For hemoglobin and platelets, these correlated with the increased incidences of anemia and thrombocytopenia noted earlier. Very few patients exhibited significant shifts in blood chemistry parameters, other than in uric acid, where the proportion of patients with significant shifts was higher in the placebo group than in the Tarceva group.
Conclusions
The combination of Tarceva with cisplatin and gemcitabine for the first-line treatment of NSCLC did not result in a survival benefit or increased quality of life. The addition of Tarceva was associated with a small increase in serious adverse events, but did not increase the incidence of treatment-related deaths. Caution (requiring special attention to the hydration of patients) should be used when Tarceva is combined with cisplatin in any future studies in other indications. Intensive PK sampling from a small group of patients did not detect any effect of Tarceva on cisplatin or gemcitabine plasma levels, or vice versa.
Publications (references, if available)
Gatzemeier U, Pluzanska A, Szczesna E. et al; Results of a Phase III trial of erlotinib (OSI-774) combined with cisplatin and gemcitabinne (GC) chemotherapy in advanced non-small cell lung cancer (NSCLC). Journal of Clinical Oncology, 2004 ASCO Annual Meeting Proceedings (Post-Meeting Edition). Vol 22, No 14S (July 15 Supplement), 2004: Abstract No.7010
Date of report
3/31/2004
|
