Evidence for Early Second-line Therapy: Docetaxel and Erlotinib

In advanced NSCLC, ASCO recommends that chemotherapy should be stopped after a maximum of 6 cycles, even in the presence of response.³³ This recommendation was based on studies by Smith et al³⁴ and Socinski et al,³⁵ who randomized patients with advanced disease to either a shorter or longer duration of the same platinum chemotherapy (3 versus 6 cycles, and 4 cycles versus treatment until progression, respectively). It is important to note that this randomization occurred at the beginning of treatment, before any benefit from the administered chemotherapy could be assessed. Although negative in their outcome, these studies revealed some significant findings: only about half of the patients were able to complete 4 cycles of treatment, and only 30% received 6 cycles due to cumulative toxicity. However, all responses occurred within the first 4 cycles, while toxicity was clearly cumulative with ongoing chemotherapy.

The realization that second-line, and even third-line therapy can extend survival in patients with advanced disease, and the FDA approval of three agents in this setting (docetaxel, pemetrexed, and erlotinib), set the stage for more innovative trial designs. Clinical researchers wanted to assess the impact of these agents when used immediately following upfront therapy and before any documented progression. This design has several advantages over the previous "duration of chemotherapy" studies. First, patients with early progression can be routed for appropriate treatment, allowing only those patients with stable or responsive tumors to be considered for continuation of therapy. This is more consistent with the clinical question at hand: whether prolongation of chemotherapy is prudent for patients who already exhibit some benefit from treatment. Additionally, using a non-cross-resistant regimen with documented efficacy in the second-line setting can lead to additional responses and, hopefully, improved global tumor control. Since the intensity of therapy is reduced to a single agent, toxicities are also expected to be modest, allowing for better treatment delivery. Studies of "early second-line therapy" have now been completed for all three agents (docetaxel, pemetrexed, and erlotinib), although published data are only available for docetaxel.

Docetaxel

After enrollment in the docetaxel study (YEAR), patients with advanced NSCLC received 4 cycles of carboplatin (AUC 5 on day 1) and gemcitabine (1000 mg/m2 on days 1, 8) with assessment of their disease every 2 cycles.³ Patients with stable or responsive disease were then randomized to docetaxel, given either immediately after completion of carboplatin and gemcitabine (immediate arm) or later at the time of documented progression (delayed arm). For both groups docetaxel was administered at a dose of 75 mg/m2 every 3 weeks for a maximum of 6 cycles. The primary end point of the study was OS from time of randomization, while PFS, RR, and quality of life (QoL) were secondary end points. It should be noted that this is the sole study where, by design, the sole difference between the two arms was the timing of therapy.

A total of 566 patients were enrolled, and 398 completed 4 cycles of front-line therapy; 309 were subsequently randomized to the immediate (n = 153) or delayed (n = 156) arm. Although 94.8% of patients in the immediate arm received docetaxel, only 62.8% of patients in the delayed arm received at least one dose of treatment. For those patients actually receiving therapy, the median dose intensity was similar, although the mean number of cycles delivered was slightly higher in the immediate arm (4.4 versus 3.8 cycles). Median PFS from randomization was significantly longer for the immediate arm than the delayed arm (5.7 versus 2.7 months; P = 0.001). Similarly, median survival time favored the immediate arm (12.3 versus 9.7 months), but this difference was not statistically significant (P = 0.08). QoL was assessed using the Lung Cancer Symptom Scale (LCSS) questionnaire at baseline and during docetaxel therapy for the immediate arm, or during observation for the delayed arm. However, QoL results were not statistically different between the two groups (P = 0.76).

There are several points worth addressing regarding data obtained from the YEAR study.

• There appears to be a real benefit from immediate docetaxel therapy in terms of PFS. Not only were the results statistically significant, but the 3-month difference is also clinically meaningful especially considering the lack of observable detriment in QoL for patients receiving immediate therapy. Similar PFS advantages for "early second-line therapy" have been observed in the study by Belani et al with the use of pemetrexed, and in the ATLAS and SATURN studies with the use of erlotinib (see below).^4,5,6 Interestingly, these studies used PFS as their primary end point.
• Immediate docetaxel therapy does not improve OS at a statistically significant level. However, the power of the study to detect the observed 2.6-month difference was only 50%, since it was originally designed to detect a 4-month difference. For patients actually receiving docetaxel, the median survival time was identical at 12.5 months. This finding suggests that overall exposure to therapy, rather than timing, may be the most important factor.
• In the delayed arm, 58 out of 156 patients did not receive docetaxel. In the majority of cases, either disease progression or symptoms were significant enough to preclude additional chemotherapy (n = 25 progressive disease; n = 16 patient or MD decision; n = 3 death). Despite the large number of patients unable to be treated at time of progression, this study should not be viewed as a comparison between second-line therapy and no therapy. Rather, it is a comparison of two different strategies: immediate non-cross-resistant therapy after completion of upfront chemotherapy versus observation with planned treatment upon progression, recognizing that only a fraction of the patients will be able to receive therapy. This point is made very clear in several recent chemotherapy trials for advanced NSCLC in which the rate of second-line therapy, which sometimes included the use of EGFR TKIs, ranged from 42-56%.
• No variables were identified as useful in predicting a patient's ability to receive docetaxel upon progression. We evaluated several baseline randomization characteristics as potential predictors of second-line docetaxel treatment in the delayed arm, including age, sex, stage (IIIB versus IV), histology, response status at randomization (complete/ partial response versus stable disease), and performance status (ECOG PS 0-1 versus 2). There was no indication that any of these variables could guide the clinician in predicting the patient's ability to receive docetaxel upon progression.

50% of patients on the delayed arm had relapsed within 2.7 months after completing front-line therapy. In my view, this statistic justifies a relatively early (4- to 6-week) restaging evaluation, following which further recommendations can be made about the frequency of scanning intervals.

Gefitinib

Hida et al reported on an interesting study at the 2008 ASCO meeting.³⁶ Patients with advanced NSCLC were randomized upfront to receive either a maximum of 6 cycles of a platinum doublet, or 3 cycles of chemotherapy with a scheduled transition to gefitinib at a dose of 250 mg daily. The study included a prespecified analysis for patients with adenocarcinoma histology. Progression-free survival was significantly prolonged for the gefitinib arm, with a HR of 0.68 (P < 0.001). Overall survival was not significantly longer in the gefitinib arm for the population as a whole, but in the 467 patients with adenocarcinoma there was a significant 1.1-month difference (HR = 0.79, P = 0.03) favoring early transition to an EGFR TKI.

Interestingly, the survival benefit was driven exclusively by the group of smokers randomized to gefitinib, who had a 3-month survival advantage over the chemotherapy-alone arm. Never-smokers had excellent survival, in excess of 20 months, although this did not differ between the two arms. This presumably reflects the activity of gefitinib as a salvage regimen in this group of patients, and suggests that early use of an EGFR TKI may be best in an unselected population.

Erlotinib

Preliminary results from two randomized phase III trials, which enrolled more than 3000 patients and used erlotinib as "early second-line therapy," showed encouraging results. SATURN was an international study that evaluated erlotinib or placebo, given immediately after initial treatment with at least 4 cycles of platinum-based chemotherapy.⁵ ATLAS was also a large global trial that utilized 4 cycles of bevacizumab in combination with the investigators' choice of a platinum-based doublet as front-line therapy.⁶ Patients without evidence of progressive disease were subsequently randomized to receive either bevacizumab and erlotinib or bevacizumab and placebo. Both studies showed an improvement in PFS for the group randomized to "early" erlotinib, which was the primary end point of the trials.

At this point we do not have survival data for either study, which are eagerly awaited. In the SATURN trial, investigators evaluated a variety of biomarkers for their predictive or prognostic value. These included EGFR gene copy number as measured by FISH, EGFR expression by immunohistochemistry, EGFR mutation, and K-ras mutation. Positive findings were seen in 48%, 84%, 11%, and 18% of evaluable samples, respectively. K-ras mutation was prognostic for worse outcome, as evidenced by a significantly inferior PFS compared to the K-ras wild-type group within the control arm. In contrast, EGFR mutation was the only predictive marker for PFS, with a HR of 0.78 for the wild-type gene compared to a HR of 0.10 for the mutant gene (P value for interaction <0.0001) Similar findings were suggested in the ATLAS trial, where the PFS HR for never-smokers, a clinical feature strongly associated with EGFR mutations, was 0.34 compared with 0.76 for ever-smokers. Collectively, these studies indicate that never smokers should be tested for EGFR mutations and, if positive, should receive erlotinib earlier rather than later, and probably even as first-line therapy.

NEXT > >