Sangha, Randeep MD*; Davies, Angela M. MD†; Lara, Primo N. Jr MD‡; Mack, Philip C. PhD‡; Beckett, Laurel A. PhD‡; Hesketh, Paul J. MD§; Lau, Derick MD‡; Li, Tianhong MD‡; Perkins, Natasha‡; Gandara, David R. MD‡
Platinum-based chemotherapy has long been the standard of care for patients with advanced non-small cell lung cancer (NSCLC). In recent years, new strategies integrating molecular-targeted therapies have modified the treatment paradigm. Inhibition of the epidermal growth factor receptor (EGFR) pathway with small molecule tyrosine kinase inhibitors (TKIs), such as erlotinib and gefitinib, was the first targeted therapeutics to demonstrate efficacy as single agents in NSCLC. A large phase III trial demonstrated a survival advantage of erlotinib compared with placebo in the second- or third-line setting.1 Importantly, this outcome was achieved in an unselected patient population with no requirement for tumor EGFR expression. Similarly, docetaxel chemotherapy is a well-established second-line therapeutic option, which has been shown to result in improved survival when compared with placebo or other chemotherapeutic agents.2
Despite preclinical data suggesting that EGFR-TKIs would have additive-to-synergistic effects when combined with an array of chemotherapeutic agents, four large randomized phase III trials combining erlotinib or gefitinib together with first-line platinum-based chemotherapy failed to increase efficacy when compared with chemotherapy alone.3–6 Multiple possible explanations for these unexpected and disappointing results have been offered, including lack of patient selection, especially in view of the subsequent discovery of activating mutations in EGFR. Alternatively, we have hypothesized a negative interaction between EGFR-TKIs and chemotherapy in cancers with wild-type EGFR, which may account for the lack of benefit of combination therapy in previous phase III trials.7 Preclinical models suggest that although EGFR-TKIs are cytotoxic in NSCLC cell lines harboring EGFR-activating mutations, inducing apoptosis through caspase-related mechanisms, they are primarily cytostatic in EGFR wild-type cell lines.8,9 We and others have shown evidence in support of sequence specificity and schedule-dependent interactions of EGFR-TKIs and chemotherapy, suggesting that strategies involving intermittent EGFR-TKI dosing between chemotherapy cycles to achieve pharmacodynamic separation might prove most efficacious.10–13 The hypothesized explanation for this phenomenon exists in the potential antagonism between targeted therapies and chemotherapy: EGFR-TKIs induce G1-phase cell-cycle arrest, which protects cells from the cytotoxic effects of cell-cycle phase-dependent chemotherapeutic agents.
To test this theory, we conducted a phase I/II study designed to assess the safety and feasibility of two different schedules intercalating intermittent erlotinib and docetaxel. On the basis of the phase I data, a single dosing schedule was used in the phase II efficacy evaluation.
PATIENTS AND METHODS
Patients with histologically or cytologically confirmed advanced solid tumors treated with any number of prior chemotherapy regimens were eligible for the phase I portion of the trial. Eligibility for the phase II portion consisted of cytologically or histologically confirmed NSCLC, treated with no greater than one previous treatment for metastatic disease. Phase II patients must have had progressive or recurrent disease after platinum-based chemotherapy. Prior chemotherapy or radiotherapy must have been completed at least 4 or 2 weeks, respectively, before study entry, and all significant previous treatment-related toxicities had to be resolved. Additional eligibility criteria included age older than 18 years, life expectancy of more than 12 weeks, and an Eastern Cooperative Oncology Group (ECOG) performance status of 0 to 2 for phase I and ECOG performance status of 0 and 1 for phase II. Adequate hematologic (absolute neutrophil count ≥1500/μl and platelet count ≥100,000/μl), renal (serum creatinine ≤1.5 mg/dl or a calculated creatinine clearance ≥50 ml/min), and hepatic function (serum bilirubin less than or equal to the institutional upper limit of normal and aspartate aminotransferases within 2.5 times the upper limit of normal) were required. Patients with asymptomatic treated brain metastasis (surgical resection or radiotherapy) were eligible if they were neurologically stable and had stopped taking steroids and/or anticonvulsants for a minimum of 4 weeks. Patients who had previously received EGFR-targeted therapy or docetaxel were excluded. Patients with clinically significant ophthalmologic abnormalities or other disorders that might increase the risk of corneal epithelial injury were excluded. Patients with preexisting peripheral neuropathy ≥grade 2 or who were pregnant were not eligible. Patients of childbearing potential were required to use a medically acceptable contraceptive. For phase II, no prior malignancy was allowed except for the following: adequately treated basal cell or squamous cell carcinoma of the skin, in situ cervical carcinoma, adequately treated stage I or II cancer in complete remission, or any other malignancy from which the patient had been disease-free for greater than 5 years. The Institutional Review Board at the University of California, Davis approved this study, which was conducted in accordance with federal and institutional guidelines.
Study Design and Treatments
The phase I dose-escalation portion of the trial was designed to determine feasibility, evaluate safety, and to recommend a phase II dose and schedule of intercalation for intermittent erlotinib and docetaxel. During the phase I portion, patients were treated on one of the two independently accruing treatment arms according to a predetermined dose-escalation schema (Figure 1). Patients were enrolled using a standard 3 + 3 design. Once accrual to the first dose level on treatment arm A was completed, treatment arm B was open for enrollment to its first dose level. Accrual was continued in an alternating “ping-pong” fashion between both arms. Although combined in a single trial, the intent was to simulate concurrently run phase I trials.
In arm A, docetaxel was administered intravenously on day 1 of a 21-day cycle and erlotinib was given weekly on days 2, 9, and 16. In arm B, docetaxel was given intravenously on day 1of a 21-day cycle and erlotinib was given on days 2 to 16. To decrease the frequency and severity of known docetaxel-induced toxicities such as hypersensitivity reaction, fluid retention, and cutaneous adverse events (AEs), all patients received dexamethasone 4 or 8 mg orally twice daily for 3 days, starting 12 to 24 hours before docetaxel infusion. Erlotinib was taken orally in the morning with 200 ml of water 1 hour before or 2 hours after meals. Treatment with both drugs continued in the absence of disease progression, provided patients were clinically benefiting and tolerating treatment. After six cycles of intermittent erlotinib and docetaxel, patients were permitted to continue with erlotinib monotherapy (Figure 1).
Three patients were treated per cohort for one cycle (21 days) with no intrapatient dose escalation. If no dose-limiting toxicities (DLTs) were recorded, treatment continued and three patients were treated in the subsequent cohort. However, if a patient developed a DLT, another three patients (for a total of six) were treated within that cohort for one cycle. If no additional DLTs were observed, dose escalation continued. Escalation was terminated when two or more patients experienced DLT attributable to study drug at any given dose level. DLT was defined as any of the following occurring within the first cycle of treatment: grade 4 thrombocytopenia; grade 3 thrombocytopenia associated with bleeding, transfusion requirement, or lasting greater than 7 days; febrile neutropenia or grade 4 neutropenia of greater than 7 days duration; grade ≥3 peripheral neuropathy; or any grade ≥3 nonhematological toxicity considered by the investigator to be clinically significant and related to study drug, except for alopecia or inadequately treated nausea, vomiting, and/or diarrhea. Toxicity was graded according to the National Cancer Institute Common Toxicity Criteria version 2.0. The maximum tolerated dose (MTD) was defined as the highest dose level at which no more than one patient experienced DLT when at least six patients were treated at that dose level and were assessable for toxicity.
On the basis of the phase I multiple-dosing portion of this study, treatment arm B dosing schedule was selected for the phase II portion of this trial. Patients received 70 mg/m2 of intravenous docetaxel on day 1 and 200 mg of oral erlotinib on days 2 through 16 of a 21-day cycle. Patients without progression could receive a maximum of six cycles of treatment and erlotinib monotherapy thereafter. Docetaxel dose reductions were required for patients with significant hematological or nonhematological toxicities in the preceding cycle. Grade ≥3 peripheral neuropathy required removal from the study. Erlotinib dose reductions were required for intolerable grade 2 or grade ≥3 cutaneous toxicity. Grade ≥2 diarrhea and nausea/vomiting despite appropriate loperamide or antiemetic use also required dose modification. Treatment delay of more than 3 weeks because of toxicity required removal of the patient from the study. Routine use of granulocyte colony-stimulating factors was permitted for the phase II portion of the study in accordance with American Society of Clinical Oncology guidelines.
Baseline imaging and hematology plus blood chemistry were obtained within 4 and 2 weeks of study entry, respectively. Baseline screening included a complete medical history, physical examination, list of concomitant medications, assessment of ECOG performance status, and body weight. Weekly complete blood counts with white blood cell differentiation were drawn, while blood chemistries were performed before each 21-day cycle. Exploratory pharmacokinetic analyses were performed in a limited subset of patients. AEs were evaluated on days 1 and 8 of the first cycle and before each subsequent cycle using the National Cancer Institute Common Toxicity Criteria version 2.0.
Tumor size was assessed through computed tomography every 42 days. RECIST were used to determine tumor response and disease progression. The primary objective of the phase II portion was to determine response rate (RR) of intermittent erlotinib combined with docetaxel when administered as second-line treatment in patients with advanced NSCLC who previously received one platinum-containing regimen. Progression-free survival (PFS), overall survival (OS), and safety were secondary objectives. Patient archival specimens and blood samples were submitted for exploratory molecular analyses.
For phase II, a one-stage design was used to test the null hypothesis that objective RR did not exceed 5% at a 0.05 level of significance. Target accrual was 38 patients, to provide 90% power to reject the null hypothesis if true objective RR was ≥20%. With a sample size of 38 patients, there was 90% certainty that at least one toxicity occurring as much as 6% time would be captured. Kaplan-Meier methods were used to determine PFS and OS.
Eighty-one patients were enrolled in this study. For phase I, 17 patients were accrued to treatment arm A and 25 patients for treatment arm B. In the phase II portion of the study, 39 patients with advanced NSCLC were treated. Baseline demographics and disease characteristics are outlined in Tables 1 and 2.
Treatment arm A.
Five patients were accrued at the original dose of 1200 mg of erlotinib weekly and 70 mg/m2 of docetaxel, with two experiencing DLTs: one fatal sepsis and one patient with pulmonary embolism. At a revised dose level 1, three patients were enrolled with one patient experiencing DLT from febrile neutropenia (Table 3). An additional three patients were accrued to dose level 1 without DLT. Dose level 2 saw two of six patients with DLTs of febrile neutropenia and grade 3 infection. Although the MTD for treatment arm A was established at the revised dose level 1 of 600 mg erlotinib weekly and 70 mg/m2 of docetaxel on day 1 of each 21-day cycle, this dose schedule was not considered feasible for phase II evaluation.
Treatment arm B.
One of three patients had a DLT of febrile neutropenia at dose level 1, and an additional three patients were enrolled at this dose level without DLT. Subsequently, dose level 2 was opened and three patients were treated without DLT. At dose level 5, two of three patients experienced DLTs of grade 3 diarrhea and febrile neutropenia. The dose level was de-escalated but, unexpectedly, further DLTs were then observed in patients at dose level 4 (grade 3 mucositis and febrile neutropenia) and dose level 3 (grade 4 infection and febrile neutropenia). Hence, the MTD for treatment arm B was established at dose level 2 of 200 mg erlotinib on days 2 through 16 with 70 mg/m2 of docetaxel day 1 of each 21-day cycle.
Thirty-seven patients were evaluable for response in the phase I portion of the trial. Twenty-one patients had NSCLC, and 16 patients had other solid tumor malignancies. Five patients were not evaluable for response and went off study after one cycle without repeat disease assessment. Of the five patients treated on arm A at the original dose level of 1200 mg erlotinib and 70 mg/m2 docetaxel, two patients with advanced NSCLC had stable disease (SD) as best response. Response data for patients with NSCLC are summarized in Table 4. In treatment arm A, six patients had SD, three each for dose levels 1 and 2. In treatment arm B, five patients had SD (one each in dose levels 1, 2, and 4; two for dose level 3), two had partial responses (PRs) (dose level 1), and a complete response was also observed in one patient (dose level 2).
Response data for patients with solid tumors other than NSCLC are summarized in Table 5. In arm A, SD was observed in patients with primary unknown (dose level 1) and prostate cancer (dose level 2). In arm B, two PRs were seen in patients with hepatocellular and bladder cancer (dose level 2), and SD was observed in patients with small-cell lung cancer (dose level 1), germ-cell malignancy (dose level 2), ovarian cancer (dose level 3), and in a patient with adenoid cystic cancer (dose level 4).
AEs for patients enrolled on treatment arms A and B were similar. Common grade ≥3 toxicities were hematologic, with neutropenia occurring most often (Table 6). Nonhematologic AEs were primarily grades 1 and 2 with rash, fatigue, and diarrhea being the most common AEs. One patient experienced grade 4 rash, and three patients experienced grade 3 diarrhea.
Nadir plasma concentrations of erlotinib and its primary metabolite (OSI-420) were assessed in five patients in arm B, before their second cycle of docetaxel (day 21) to ascertain if the goal of pharmacodynamic separation was achieved.14 With this erlotinib dosing schedule of days 2 through 16, both erlotinib and OSI-420 were either undetectable (below limit of quantitation <1.00 ng/ml) or far below therapeutic levels (estimated 183 ng/ml for wild-type EGFR and 66 ng/ml for the L858R EGFR mutation) in four of five cases sampled (Table 7).15 These subtherapeutic plasma concentrations prevent potential antagonism between erlotinib and subsequent docetaxel dosing, supporting the concept of pharmacodynamic separation. We have previously demonstrated analogous findings using a similar intercalated design with erlotinib and pemetrexed.16 In the one outlier (erlotinib level 320 ng/ml), continued dosing of erlotinib beyond the protocol-specified stop date is suspected.
Treatment arm B was selected for further investigation in the phase II portion of the study given the feasibility of administration, promising number of responses, and achievement of pharmacodynamic separation. Erlotinib 200 mg was given daily for days 2 through 16, and docetaxel 70 mg/m2 was administered on day 1 of each 21-day cycle to 39 patients with advanced NSCLC who had progressed after a platinum-containing regimen.
Best response to therapy is shown in Table 8. The overall RR was 28.1%, and the disease-control rate was 64.1%. Although most of patients did not have tissue available for EGFR mutation analysis, we tested circulating tumor DNA in plasma for the presence of EGFR mutations.17 Two of the patients with complete response and four of nine patients with PR possessed EGFR mutations. Median PFS was 4.1 months (Figure 2). The 1-year PFS was 17.9%. Median OS was 18.2 months, and survival rates at 1 and 2 years were 58.9% and 38.5%, respectively (Figure 3).
Table 9 shows grade ≥3 treatment-related AEs for the subset of patients treated at the phase II dose (n = 39). Anemia and thrombocytopenia were common, but these were predominantly grade 1 or 2 toxicities with only one patient experiencing grade 3 thrombocytopenia. Neutropenia was common and occurred in 46% of patients and tended to be a grade 3 or 4 toxicity (36%); four patients (10%) developed febrile neutropenia. The most common nonhematological toxicities (grade ≥1) were diarrhea (79%), rash (77%), and fatigue (69%). Although 18% of patients had grade ≥3 diarrhea despite optimal preventative measures with loperamide, supplemental fluids, and other supportive measures, no patient experienced a grade ≥3 rash. The most common AEs resulting in dose reduction were diarrhea or intolerable grade 2 rash. There were no treatment-related deaths on the phase II portion of this study.
Chemotherapy plays a crucial role in the treatment of most stages of NSCLC. Likewise, agents targeting the EGFR, particularly the TKIs, erlotinib and gefitinib, have proven their importance in the treatment paradigm for advanced NSCLC. Erlotinib is approved for second- and third-line therapies, and, now, in the era of personalized medicine, first-line EGFR-TKI treatment is considered a standard of care for patients with cancers possessing an EGFR mutation.1,18–20 Although the appeal of combining traditional cytotoxic chemotherapy with biological therapeutics is understandable, four large randomized phase III trials of chemotherapy with or without an EGFR-TKI in patients with unselected advanced NSCLC did not demonstrate improvement in clinical outcomes.3–6 Moreover, combinations of chemotherapy and other “targeted” small-molecule inhibitors have generally been unsuccessful in NSCLC. In contrast, monoclonal antibodies such as bevacizumab and cetuximab have shown improved efficacy when combined with chemotherapy, perhaps because of inherent differences in mechanisms of action.21–23
In this study, we report a phase I/II trial exploring a combination strategy based on pharmacodynamic parameters together with supporting preclinical data. We and others have previously reported that pretreatment with erlotinib can result in a G1 arrest, abrogating the cytotoxicity of cell-cycle-specific chemotherapy.8,24 In our studies, a sequence of M-phase-specific docetaxel followed by erlotinib exhibited the greatest levels of apoptosis and was superior to the reverse sequence (erlotinib followed by docetaxel). We hypothesized that to extend these observations to the clinic, an erlotinib dose schedule intercalating intermittent delivery between docetaxel cycles would attain pharmacodynamic separation, abrogating negative cell-cycle-specific interactions of these agents.
Two erlotinib dose schedules were used in the phase I portion of this trial. Treatment arm A used high-dose erlotinib administered weekly with docetaxel. Weekly erlotinib at doses ≥600 mg result in high peak plasma levels within 4 hours.25 With a half-life of 36 hours, plasma levels from weekly therapy would be expected to be undetectable before the end of the 8-day intercalated dosing interval, allowing pharmacodynamic separation from subsequent docetaxel administration. In our trial, patients in arm A were unable to tolerate a weekly dose higher than 600 mg together with docetaxel 70 mg/m2, a cumulative erlotinib dose considerably lower than that which could be given in arm B (MTD, erlotinib 200 mg days 2 through 16 and docetaxel 70 mg/m2). Thus, the arm B schedule was chosen for further phase II efficacy evaluation.
For phase II, 39 patients with advanced NSCLC who had progressed after a platinum-containing regimen were treated with erlotinib 200 mg days 2 through 16 and docetaxel 70 mg/m2 day 1 on a 21-day cycle. Overall RR was 28.1% and disease-control rate was 64.1%. Encouragingly, median PFS was 4.1 months and OS was 18.2 months. In comparison, the BR.21 phase III trial demonstrated that patients with advanced NSCLC treated with erlotinib monotherapy after failing one or two prior chemotherapy regimens had a RR of 8.9% and median PFS and OS of 2.2 months and 6.7 months, respectively.1 Docetaxel, as second-line therapy, showed a median survival of 7.0 months in the TAX 317 trial.2 Although our results cannot be directly compared with the phase III trials described earlier in the text, the response and survival outcomes are favorable.
As previously described, EGFR-activating mutations were identified by means of an allele-specific polymerase chain reaction assay (Scorpion-amplification refractory mutation system) of circulating DNA in plasma in some patients on this trial.17 The presence of an EGFR mutation correlated significantly with improved tumor response and PFS in patients with NSCLC treated with intermittent erlotinib and docetaxel. However, the use of docetaxel in this intercalated schedule did not seem to reduce erlotinib activity in patients possessing EGFR mutations. In addition, patients with wild-type EGFR also showed benefit.
In conclusion, this study describes clinical application of a model of pharmacodynamic separation designed to overcome hypothesized negative interactions of concurrent administration of erlotinib and chemotherapy. Others have recently reported using this approach to administer EGFR-TKIs and other chemotherapeutic regimens. For example, in the FASTACT trial, Mok et al.26 demonstrated improved RR and PFS in patients receiving intercalated gefitinib plus gemcitabine-carboplatin versus those randomized to chemotherapy alone. A sequential approach of administering chemotherapy before therapy with an EGFR-TKI is a potential alternative design of overcoming antagonism. One such study, SATURN, showed an improvement in survival in an unselected population of patients demonstrating disease control after treatment with four cycles of platinum agent-based chemotherapy followed by erlotinib. Whether an intercalated or sequential model is more efficacious in achieving pharmacodynamic separation will require further exploration.27
Considering the inherent myelosuppression associated with docetaxel in previously treated patients, in our own studies in second-line therapy for NSCLC, we are currently evaluating a similar strategy with intercalated erlotinib and pemetrexed.16 Further studies exploring pharmacodynamic separation are ongoing.
1. Shepherd FA, Rodrigues Pereira J, Ciuleanu T, et al. Erlotinib in previously treated non-small-cell lung cancer. N Engl J Med
2. Shepherd FA, Dancey J, Ramlau R, et al. Prospective randomized trial of docetaxel versus best supportive care in patients with non-small-cell lung cancer previously treated with platinum-based chemotherapy. J Clin Oncol
3. Giaccone G, Herbst RS, Manegold C, et al. Gefitinib in combination with gemcitabine and cisplatin in advanced non-small-cell lung cancer: a phase III trial–INTACT 1. J Clin Oncol
4. Herbst RS, Giaccone G, Schiller JH, et al. Gefitinib in combination with paclitaxel and carboplatin in advanced non-small-cell lung cancer: a phase III trial–INTACT 2. J Clin Oncol
5. Herbst RS, Prager D, Hermann R, et al. TRIBUTE: a phase III trial of erlotinib hydrochloride (OSI-774) combined with carboplatin and paclitaxel chemotherapy in advanced non-small-cell lung cancer. J Clin Oncol
6. Gatzemeier U, Pluzanska A, Szczesna A, et al. Phase III study of erlotinib in combination with cisplatin and gemcitabine in advanced non-small-cell lung cancer: the Tarceva Lung Cancer Investigation Trial. J Clin Oncol
7. Gandara DR, Davies AM, Gautschi O, et al. Epidermal growth factor receptor inhibitors plus chemotherapy in non-small-cell lung cancer: biologic rationale for combination strategies. Clin Lung Cancer
8. Davies AM, Ho C, Lara PN Jr, et al. Pharmacodynamic separation of epidermal growth factor receptor tyrosine kinase inhibitors and chemotherapy in non-small-cell lung cancer. Clin Lung Cancer
9. Tracy S, Mukohara T, Hansen M, et al. Gefitinib induces apoptosis in the EGFRL858R non-small-cell lung cancer cell line H3255. Cancer Res
10. Mahaffey CM, Davies AM, Lara PN Jr, et al. Schedule-dependent apoptosis in K-ras mutant non-small-cell lung cancer cell lines treated with docetaxel and erlotinib: rationale for pharmacodynamic separation. Clin Lung Cancer
11. Piperdi B, Ling YH, Kroog G, et al. Schedule-dependent interaction between epidermal growth factor imhibitors (EGFR1) and G2/M blocking chemotherapeutic agents (G2/MC) on human NSCLC cell lines in vitro. J Clin Oncol
12. Solit DB, She Y, Lobo J, et al. Pulsatile administration of the epidermal growth factor receptor inhibitor gefitinib is significantly more effective than continuous dosing for sensitizing tumors to paclitaxel. Clin Cancer Res
13. Li T, Ling YH, Goldman ID, et al. Schedule-dependent cytotoxic synergism of pemetrexed and erlotinib in human non-small cell lung cancer cells. Clin Cancer Res
14. Hamilton M, Wolf JL, Rusk J, et al. Effects of smoking on the pharmacokinetics of erlotinib. Clin Cancer Res
15. Carey KD, Garton AJ, Romero MS, et al. Kinetic analysis of epidermal growth factor receptor somatic mutant proteins shows increased sensitivity to the epidermal growth factor receptor tyrosine kinase inhibitor, erlotinib. Cancer Res
16. Davies AM, Ho C, Beckett L, et al. Intermittent erlotinib in combination with pemetrexed: phase I schedules designed to achieve pharmacodynamic separation. J Thorac Oncol
17. Mack PC, Holland WS, Burich RA, et al. EGFR mutations detected in plasma are associated with patient outcomes in erlotinib plus docetaxel-treated non-small cell lung cancer. J Thorac Oncol
18. Mok TS, Wu YL, Thongprasert S, et al. Gefitinib or carboplatin-paclitaxel in pulmonary adenocarcinoma. N Engl J Med
19. Maemondo M, Inoue A, Kobayashi K, et al. Gefitinib or chemotherapy for non-small-cell lung cancer with mutated EGFR. N Engl J Med
20. Zhou C, Wu YL, Chen G, et al. Efficacy results from the randomised phase III OPTIMAL (CTONG 0802) study comparing first-line erlotinib versus carboplatin plus gemcitabine, in chinese advanced non-small-cell lung cancer patients with EGFR activating mutations. Ann Oncol
21. Sandler A, Gray R, Perry MC, et al. Paclitaxel-carboplatin alone or with bevacizumab for non-small-cell lung cancer. N Engl J Med
22. Manegold C, von Pawel J, Zatloukal P, et al. BO17704 (AVAIL): a phase III randomised study of first-line bevacizumab combined with cisplatin/gemcitabine in patients with advanced or recurrent non-squamous, non-small cell lung cancer. Ann Oncol
23. Pirker R, Pereira JR, Szczesna A, et al. Cetuximab plus chemotherapy in patients with advanced non-small-cell lung cancer (FLEX): an open-label randomised phase III trial. Lancet
24. Perez-Soler R, Piperdi B, Haigentz M, et al. Determinants of sensitivity to the EGFR TK inhibitor erlotinib in a panel of NSCLC cell lines. J Clin Oncol
25. Milton DT, Azzoli CG, Heelan RT, et al. A phase I/II study of weekly high-dose erlotinib in previously treated patients with nonsmall cell lung cancer. Cancer
26. Mok TS, Wu YL, Yu CJ, et al. Randomized, placebo-controlled, phase II study of sequential erlotinib and chemotherapy as first-line treatment for advanced non-small-cell lung cancer. J Clin Oncol
27. Cappuzzo F, Ciuleanu T, Stelmakh L, et al. Erlotinib as maintenance treatment in advanced non-small-cell lung cancer: a multicenter, randomised, placebo-controlled phase 3 study. Lancet Oncol