Cisplatin and Paclitaxel for Non–Small-Cell Lung Cancer: The European Experience

Introduction

Lung cancer is the most common malignant disease in the western world. Although the incidence is stabilizing or even decreasing in men in this area, the number of women affected is increasing. In eastern Europe, China, and the developing countries the incidence in both genders is rapidly increasing.

Treatment of small-cell lung cancer has developed during the past 25 years from local therapy, surgery, and radiotherapy, to systemic treatment, combination chemotherapy, and recently the addition of radiotherapy for disease limited to the thorax. For non–small-cell lung cancer the treatment has been surgery for the earlier stages (I and II) and radiotherapy for stage III. For patients with disseminated disease the development of treatment has not followed the same route as for small-cell lung cancer due to the lack of clearly active chemotherapeutic agents. However, for stage IV disease, discussion of the role of systemic therapy has recently changed.

In a meta-analysis of a large number of studies comparing one treatment modality with that same treatment plus chemotherapy, it was demonstrated that the addition of combination chemotherapy to best supportive care resulted in a survival improvement at 1 year of 10% to 25% and an improvement of the median survival from 4 to 6 months.[1] Still this benefit is regarded as very small and serious doubts remain whether this was sufficient to consider combination chemotherapy as standard treatment for disseminated non–small-cell lung cancer. One of the main reasons for these doubts was the important role of cisplatin(Drug information on cisplatin) (Platinol) demonstrated in this meta-analysis. It remained questionable whether the survival benefit was offset by the toxicity of the chemotherapy, especially of cisplatin.

At the beginning of the 1990s new classes of drugs became available for clinical evaluation: taxanes, topo-I-isomerases, and a new antimetabolite.[2] In a number of phase II studies activity was demonstrated against non–small-cell lung cancer. The activity of paclitaxel(Drug information on paclitaxel) (Taxol) was especially impressive, with a response rate > 20% and a 1-year survival rate of > 40%.[3,4]

Combining paclitaxel with cisplatin was a logical step in the development of chemotherapy for non–small-cell lung cancer; hence a number of studies were started in Europe.

Phase I/II Studies

In a phase I study Rowinsky et al[5] reported that administration of cisplatin followed by a 24-hour infusion of paclitaxel was more myelotoxic than the reverse (paclitaxel then cisplatin). The recommended doses were 75 mg/m² for cisplatin and 130 to 170 mg/m² for paclitaxel. The same group investigated the maximum tolerated doses of both drugs when administered with granulocyte colony-stimulating factor. Dose-limiting toxicity was peripheral neuropathy at a 300-mg/m² dose of paclitaxel with 75 mg/m² of cisplatin.[6]

In a study by Klastersky et al[7] the dose-limiting toxicity was defined in a study starting with cisplatin at a dose of 100 mg/m² and a 3-hour paclitaxel infusion at 135 mg/m² every 3 weeks. It was possible to increase the doses to 120 mg/m² of cisplatin and 200 mg/m² of paclitaxel without dose-limiting acute toxicity. After 3 to 4 courses neurotoxicity appeared to be dose-limiting; it became clear that prolonged treatment with this combination was impossible. In 8 of 17 patients a response was seen, with an overall response rate of 47%.

The results of this study were more or less confirmed by others (Table 1): in all the phase II studies in Europe combining cisplatin and paclitaxel, the response rate was around 40%.[8-10] Based on these results it was necessary to evaluate the combination of cisplatin and paclitaxel in a randomized study, comparing it with one of the old standard regimens.

Phase III Study

The Lung Cancer Cooperative Group of the European Organization for Research and Treatment of Cancer (EORTC-LCCG) started relatively late with the evaluation of combination chemotherapy in non–small-cell lung cancer. In a randomized phase II study the role of teniposide(Drug information on teniposide) (Vumon) was evaluated and the difference in activity between a 1-day schedule and a 3-day schedule was proven.[11] This is comparable to the activity of etoposide(Drug information on etoposide) in small-cell lung cancer in which prolonged treatment gave a much higher response rate than a 1-day infusion with etoposide. In the EORTC study the combination of teniposide with cisplatin demonstrated a higher activity than teniposide alone. The best survival was achieved in the group treated with cisplatin and teniposide using a 3-day schedule as opposed to the 1-day schedule.

Based on these results the EORTC-LCCG considered cisplatin and teniposide their standard treatment. For any phase III study the group recommended that the standard arm should be teniposide and cisplatin and the experimental arm should include the most active old single agent—cisplatin. At that time the possibility of combining cisplatin with a new drug had not been extensively tested. Data on the combination of cisplatin and paclitaxel were among the earliest of the possible new combinations; therefore the group decided to compare this promising new combination with their standard regimen of cisplatin and teniposide. The doses of both drugs in this study were based on the study by Klastersky et al.[7] In a randomized phase II study both combinations were tested. At that time toxicity of evaluated patients in both regimens was considered acceptable and activity was confirmed.[12] As initially planned the study went on as a phase III study.[13]

Inclusion criteria were standard, excluding previous malignancies, prior chemotherapy, brain metastases, poor performance status, organ failure, and age > 75 years. All patients had measurable or evaluable disease. Response was assessed every two cycles by the optimal method to define it, and toxicity was scored during and after each cycle.

In 2.5 years 332 patients were included and 15 patients were not eligible. The sample size was based on an increase of the median survival from 7 months in the standard arm to 10 months in the experimental arm. Treatment consisted of 80 mg/m²of cisplatin on day 1 in both arms, and 175 mg/m² of paclitaxel prior to cisplatin on day 1 in the experimental arm and 100 mg/m² of teniposide on days 1, 3, and 5 in the standard arm. In both arms six cycles were planned every 3 weeks.

Reasons for delaying chemotherapy or dose-adjustment were standard. The median age of the patients in both arms was 58.5 years; prognostic factors such as weight loss, prior therapy, performance status, and stage were balanced between the two arms. Performance status was in general good; 11% had a performance status of 2, 55% a performance status of 1, and around 34% a performance status of 0. Weight loss > 5% occurred in 27% of the patients. Stage IV was noted in 61.5% of the patients. Adenocarcinoma was the most common histologic type, in slightly more than half of the patients.

The response rate was significantly higher in the cisplatin and paclitaxel arm (41%) vs the teniposide and cisplatin arm (28%). Despite this, median survival was not different between the two arms at 9.7 and 9.9 months, respectively. The survival rate after 1 and 2 years was also comparable (Table 2). However, toxicity, especially myelotoxicity, was more severe in the teniposide and cisplatin arm; in the cisplatin and paclitaxel patients neurotoxicity was far more common, although it did not often lead to dose-reduction or discontinuation of the administration of paclitaxel (Table 3). Overall the number of courses was higher in the cisplatin-paclitaxel group and toxicity was less.

In a subgroup of the patients a quality-of-life measurement was performed, reporting a better quality of life during the first period of the treatment. Later on this difference disappeared with a rapidly dropping number of forms available for analysis in both arms.

Discussion

Since the demonstration of a survival benefit for patients with stages IIIB and IV disease treated with combination chemotherapy, the use of chemotherapy has attracted more interest. Even physicians who previously had very negative attitudes toward chemotherapy for this disease are now willing to consider this treatment for selected patients. Relatively young patients with good prognostic factors are especially considered good candidates.

Furthermore, one should keep in mind that the results of the meta-analysis are based on what are nowadays called old chemotherapy regimens. With the introduction of the new agents such as paclitaxel, docetaxel(Drug information on docetaxel) (Taxotere), irinotecan(Drug information on irinotecan) (Camptosar), navelbine (Vinorelbine), and gemcitabine(Drug information on gemcitabine) (Gemzar), the possibilities of chemotherapy are probably changing. The higher activity of each of these new agents compared to the older agents has enlarged the expectations of chemotherapy considerably. Furthermore the toxicity was found to be acceptable and certainly not worse than with the older agents.

In a number of phase II studies in previously untreated patients the activity of these agents has been confirmed. One of the most striking observations in patients treated with paclitaxel as a single agent was the high percentage of 1-year survivors. A logical step in the further development is the use of the active new agent with the other active agents. At this stage most trials are focusing on the evaluation of the new agent in combination with what is considered to be the cornerstone of chemotherapy for non–small-cell lung cancer—cisplatin. Standard treatment regimens from the 1980s are used for comparison, for instance, cisplatin and etoposide or cisplatin and vindesine(Drug information on vindesine). A convenient schedule of cisplatin and paclitaxel was found in a number of phase II studies and considered suitable for evaluation in a multicenter randomized study. The EORTC-LCCG found the combination to be active in the phase II part of the study and decided to continue accrual for the planned phase III study. The hope of gaining survival benefit was unfortunately not fullfilled. Despite a significantly higher response rate, the control arm was comparable with regard to median survival times and 1- and 2-year survival rates. The experimental regimen behaved as expected with a response rate, median survival time, and 1- and 2-year survival rate arm as found in the phase II studies. The reason why the standard regimen resulted in a longer survival time than in the study[11] in which it was initially evaluated is unclear. It is possible that the selection of patients was more restricted with relatively fewer patients with a performance status of 2.

However, the toxicity and quality of life were worse in the standard treatment arm. Despite a reduction of the dose from 120 mg/m2 to 100 mg/m2 in comparison with the previous randomized phase II study, hematologic toxicity was still rather severe with a high incidence of grades III/IV myelotoxicity, infections, and red blood cell transfusions. The toxicity of the experimental arm was as expected, with neurotoxicity and myalgia and arthralgia shortly after the infusion as the most important toxicities.

The favorable toxicity profile and quality-of-life advantage is the reason for the EORTC-LCCG to consider the cisplatin and paclitaxel arm as the new standard arm. Subsequently the LCCG is planning a new study in which the main theme is to try not to use the most toxic chemotherapeutic agent—cisplatin—but to combine two of the new and active agents. To date, the combination of 200 mg/m² of paclitaxel on day 1 with 1,000 mg/m² of gemcitabine on days 1 and 8 has been tested. Myelotoxicity was acceptable and activity was found, although the response rate is probably somewhat lower than expected. To date, an additive effect has been found.

Combining paclitaxel with other new agents is a logical step toward reducing toxicity and developing a simple outpatient regimen without the need for time-consuming hydration procedures to prevent neurotoxicity. The use of these relatively expensive new agents will become cost-effective if hospitalization for the administration of chemotherapy is not necessary. Therefore schedules with a shorter infusion time for paclitaxel will be more than welcome.