The Emerging Role of Paclitaxel Plus Carboplatin in Non–Small-Cell Lung Cancer
The Emerging Role of Paclitaxel Plus Carboplatin in Non–Small-Cell Lung Cancer
ABSTRACT: The activity and toxicity profiles of carboplatin (Paraplatin) and paclitaxel (Taxol) used as single agents in non–small-cell lung cancer made them logical agents for study in combination therapy. Once preliminary trials demonstrated that both agents could be combined safely in full doses, phase II and III trials of the combination were started. The authors review the results of these trials of paclitaxel plus carboplatin in advanced non–small-cell lung cancer, in combination with chest radiotherapy in stage III disease, and as neoadjuvant therapy in stage I and II non–small-cell lung cancer.[ONCOLOGY 13(Suppl 4):35-41, 1999]
Lung cancer is the leading causeof cancer death in both men and women in the United States. The 160,100 deaths predicted for 1998 represent 28% of all cancer deaths. The cure rate is extremely low (14%) because most patients have metastatic disease at the time of presentation, and because systemic therapies are unable to cure advanced metastatic disease. Attitudes toward lung cancer therapy began to change when cisplatin-based therapies were shown to prolong survival,[2,3] relieve symptoms, and improve quality of life at acceptable medical costs in advanced non–small-cell lung cancer.
Although cisplatin-based therapy accomplished the results described above, there was debate regarding its general utility because of its toxicity profile, inconvenience, and marginal therapeutic gains. Carboplatin (Paraplatin) has gained increasing usage in place of cisplatin in non–small-cell lung cancer because of its convenient administration, low toxicity profile, and equivalent or superior efficacy compared to cisplatin.
Paclitaxel (Taxol) as a single agent was shown to have efficacy equivalent or superior to that of cisplatin and carboplatin. When given as a short infusion, paclitaxel can be given on an outpatient basis with relatively mild toxicity. The activity and toxicity profiles of carboplatin and paclitaxel in non–small-cell lung cancer made them logical agents for combination therapy. Preliminary trials showed that both agents could be combined safely in full doses.[9-14] This led to phase II and III trials of the combination in advanced non–small-cell lung cancer, to combination with chest radiotherapy in stage III disease, and to neoadjuvant therapy in stage I and II non–small-cell lung cancer.
Carboplatin was introduced into clinical trials in advanced non–small-cell lung cancer because it could be administered easily as a short, outpatient infusion; because it lacked the toxicity and nephrotoxicity of cisplatin; and because it produced far less nausea and vomiting compared to cisplatin. Since its introduction, several randomized trials have compared carboplatin, alone or in combination, to cisplatin-based combinations. The results of such trials are summarized in Table 1. The Eastern Cooperative Oncology Group compared single-agent carboplatin to several cisplatin-based combinations. Patients not responding to carboplatin could receive a cisplatin combination after two cycles. As shown in Table 1, the 9% response rate to carboplatin was lower than the 13% to 20% response rates for cisplatin combinations. However, the toxicity rate was significantly lower and the survival was significantly longer in patients receiving carboplatin. The European Organization for Research and Treatment of Cancer compared a combination of etoposide and cisplatin to a combination of etoposide and carboplatin. The dose of carboplatin in this trial (325 mg/m²) was low and was not based on renal function. Nonetheless, the efficacy of the two regimens was similar, whereas the toxicity was considerably less in the carboplatin arm.
More recently, a comparative randomized trial using a higher dose of carboplatin (500 mg/m²) was completed. In this trial, mitomycin-C and vindesine were combined with either cisplatin (120 mg/m²) or carboplatin (500 mg/m²). The efficacy of the carboplatin arm was superior to that of the cisplatin arm both in response rate and survival. Myelosuppression in the two arms was similar although the rate of nausea, vomiting, ototoxicity, and nephrotoxicity remained lower in the carboplatin arm. These trials provide a strong rationale for the use of carboplatin in combination therapy trials in advanced non–small-cell lung cancer.
Paclitaxel was introduced into lung cancer therapy in the early 1990s and rapidly became established as one of the most effective agents for the disease. The earliest trials used a 24-hour infusion of high-dose paclitaxel (200 to 250 mg/m²).[18,19] This dose was given on an inpatient basis and required both granulocyte-colony stimulating factor (G-CSF) and combination premedication with diphenhydramine, cimetidine, and prednisone. The response rates to these 24-hour infusions are summarized in Table 2,[18-20] along with a summary of the results of single-agent paclitaxel given as a 3-hour infusion and as a 1-hour infusion at 3-week intervals.[21-25] This table also shows the results of a single trial in which the paclitaxel was given weekly. The response rates appear to be quite similar with each schedule. In each of these trials, the survival results were also similar with 1-year survival rates of approximately 40% in each trial. The results of the weekly schedule gave the highest response rate and the longest survival. It also gave the greatest dose intensity. However, this was a single-institution trial and additional studies with this regimen are necessary. There were no phase II studies with a 96-hour infusion in untreated patients, but a single study in previously treated patients showed no responses.
Long-Infusion (24-h) Paclitaxel
Studies with long-infusion paclitaxel (24 hours) plus carboplatin are summarized in Table 3. The study of Langer et al escalated doses in each patient, starting from 135 mg/m² and escalating to 215 mg/m². The study of Johnson et al gave doses of 135 mg/m² or 175 mg/m² to different groups of patients. The study of Belani et al escalated doses of both paclitaxel and carboplatin. The overall response rate in these studies was 46% with the highest response rate (61%) seen in the escalating dose study of Langer et al. This study required inpatient administration of the drugs and routine G-CSF administration due to the high rates of grade 4 neutropenia. The average median survival was 46 weeks in these studies and the average 1-year survival rate was 43%.
Short-Infusion 1- or 3-h Paclitaxel
A summary of United States studies combining 1-hour or 3-hour infusions of paclitaxel with carboplatin is provided in Table 4.[12-14,28-34] The studies of Bunn, Natale, and Rowinsky were phase I dose-escalation studies. In these studies, paclitaxel doses below 175 mg/m² were associated with lower response rates and shorter survival.[12-14] Each of these three studies recommended paclitaxel at a dose of 225 mg/m² and carboplatin (area under the concentration-time curve of 6 [AUC in mg/mL · min]) for phase II and III studies. Surprisingly, these studies found a much lower rate of thrombocytopenia than expected. This was not due to a pharmacokinetic interaction but rather appeared to be due to a platelet-sparing effect of paclitaxel. Grade 4 neutropenia occurred in only a minority of patients given these doses and febrile neutropenia was rare. G-CSF was not administered, and all therapy was given on an outpatient basis.
Some of the paclitaxel plus carboplatin combination studies with short paclitaxel infusion times gave lower doses of paclitaxel (175 mg/m²). These studies had slightly lower response rates and shorter survival compared to those with higher paclitaxel doses (200 mg/m²). There is an ongoing randomized trial in Greece to determine if 225 mg/m² of paclitaxel infused over 3 hours is preferred over 175 mg/m² over 3 hours. Preliminary results show superiority for the higher dose. One study gave higher doses of carboplatin (AUC of 9 or 11). This study reported greater toxicity without an apparent advantage in response or survival rates. Finally, one study gave the two drugs every other week. This study reported a low response rate with considerable toxicity so it cannot be recommended.
The overall response rate in Table 4 (35%) and the 1-year survival rates 42% to 50% (average 45%) compare favorably to studies with paclitaxel combined with cisplatin, and with studies employing longer paclitaxel infusions with carboplatin (Table 3).
Due to the low rates of toxicity, the convenience, and the effectiveness of this combination, it was widely adopted into community practice. Cooperative groups selected this combination to compare to their previous standard. The ongoing cooperative group trials in advanced non–small-cell lung cancer are summarized in Table 5. This combination was also incorporated into combined-modality studies in stage III non–small-cell lung cancer with radiotherapy and stages I and II non–small-cell lung cancer with surgery.