Phase 2 study of irinotecan and paclitaxel in patients with recurrent or refractory small cell lung cancer




Patients with extensive stage small cell lung cancer (SCLC) who develop disease progression with standard cisplatin-based therapy are reported to have a poor overall prognosis. Irinotecan and paclitaxel are active as single agents and exhibit preclinical synergy in SCLC cell lines. A phase 2 study was conducted to evaluate this combination in patients with recurrent or refractory SCLC.


Patients with SCLC who progressed with 1 prior chemotherapy regimen and had measurable disease present; an Eastern Cooperative Oncology Group (ECOG) performance status (PS) of 0 to 2; and adequate bone marrow, hepatic, and renal function were included in the study. Paclitaxel (at a dose of 75 mg/m2) and irinotecan (at a dose of 50 mg/m2) were administered intravenously on Days 1 and 8 of each 3-week treatment cycle. Therapy was continued until disease progression or unacceptable toxicity. The target response rate of interest was ≥30%.


A total of 55 patients were enrolled, 51 of whom received at least 1 dose of therapy. The majority of the patients had an ECOG PS of 0 or 1 (96%). A median of 3 cycles of treatment was administered, and 15 patients received ≥6 cycles. Seventeen patients experienced toxicity of grade 3 or higher (neutropenia in 8 patients and fatigue in 5 patients). The overall response rate was 21%. The median survival was 25.4 weeks, and the 1-year survival rate was 22%.


The regimen of irinotecan and paclitaxel was found to be tolerated well in patients with recurrent or refractory SCLC. Although modest anticancer activity was noted, the efficacy failed to meet the primary endpoint of interest. Cancer 2010. © 2010 American Cancer Society.

Small cell lung cancer (SCLC) accounts for approximately 15% of all cases of lung cancer diagnosed in the United States. The majority of the patients have extensive stage disease at the time of diagnosis.1 The standard treatment for patients with extensive stage SCLC is systemic chemotherapy with the combination of platinum and etoposide. This regimen is associated with a median survival of approximately 8 to 10 months and a response rate of 50% to 80%.2, 3 Despite initial responsiveness to chemotherapy, SCLC is categorized by rapid progression to a refractory disease. Recurrent disease is categorized as either sensitive or refractory based on the outcome of first-line chemotherapy. Patients who progress ≥90 days after completion of first-time chemotherapy are considered to have sensitive disease. Patients who progress while receiving first-line therapy or within 90 days of completion are believed to have refractory SCLC. The outcome for subsequent therapy varies between patients with sensitive and refractory SCLC.

Topotecan, a novel topoisomerase inhibitor, is the only agent approved by the US Food and Drug Administration for the treatment of patients with recurrent SCLC, and its use is restricted to patients with sensitive disease recurrence. This was based on a phase 3 study that compared the efficacy of topotecan with that of a combination regimen comprised of cyclophosphamide, doxorubicin, and vincristine (CAV).4 The response rate and overall survival were similar for the 2 treatment arms. However, improvement in symptoms occurred in a higher percentage of patients with topotecan than CAV. Despite the modest benefits with second-line therapy, the overall prognosis for patients with recurrent SCLC remains poor.

Irinotecan is a semisynthetic camptothecin that exhibits anticancer activity against a variety of malignancies. It is approved for the treatment of patients with metastatic colorectal cancer. Several studies have demonstrated anticancer activity with irinotecan in patients with SCLC.5-7 In preclinical studies, irinotecan exhibits synergistic or supra-additive anticancer effects in combination with paclitaxel.8, 9 Furthermore, the favorable anticancer effects appeared to be sequence specific to the administration of paclitaxel initially and then followed by irinotecan. Paclitaxel and irinotecan are also associated with a nonoverlapping toxicity profile when administered on a weekly schedule. The optimal dose and schedule of irinotecan and paclitaxel were established in a phase 1 study reported by Murren et al.10 When administered on Days 1 and 8 of each 3-week cycle, paclitaxel and irinotecan at doses of 75 mg/m2 and 50 mg/m2, respectively, were found to be tolerated well. Furthermore, there were no significant pharmacokinetic interactions between these 2 agents. Neutropenia was the dose-limiting toxicity, with diarrhea being the most common nonhematologic toxicity.

The known anticancer activity of irinotecan and paclitaxel in SCLC, the favorable interactions as a combination, and their nonoverlapping toxicity profile provided the basis for this phase 2 study in patients with refractory or recurrent SCLC.


Patient Eligibility

Patients with histologic or cytologic confirmation of SCLC were eligible for the study. Other salient eligibility criteria included disease progression with 1 prior chemotherapy, the presence of measurable disease, an Eastern Cooperative Oncology Group (ECOG) performance status (PS) of 0 to 2, and age >18 years. Qualifying laboratory criteria were as follows: leukocyte count ≥3000/μL, absolute neutrophil count (ANC) ≥1500/μL, platelet count ≥100,000/μL, serum total bilirubin ≤ institutional upper limit of normal (ULN), serum transaminases ≤2.5 × ULN, and serum creatinine ≤1.5 times ULN. At least 4 weeks should have elapsed since prior chemotherapy and 2 weeks from prior radiotherapy. All patients were required to sign an informed consent form. Patients who had received prior therapy with paclitaxel and irinotecan were excluded. The presence of untreated brain metastasis was permitted if the patient was asymptomatic from the brain lesions. Symptomatic patients should have received prior radiotherapy before the initiation of study treatment. Patients with a history of Gilbert syndrome and peripheral neuropathy of severity greater than grade 1 were excluded. In addition, patients with uncontrolled concurrent illness, major surgery within 2 weeks before registration, pregnancy, and lactation were also excluded from the study. The study protocol was approved by the institutional review board at the University of Pittsburgh.

Treatment Plan

Paclitaxel was administered at the dose of 75 mg/m2 over 60 minutes on Days 1 and 8 of every 21–day cycle, and irinotecan was administered at the dose of 50 mg/m2 over 90 minutes after the completion of paclitaxel on Days 1 and 8 of each cycle. All treatments were administered in the outpatient setting. The premedication regimen comprised of a 5-hydroxytryptamine receptor antagonist, dexamethasone, diphenhydramine, and a histamine-2 receptor antagonist. Patients who experienced cholinergic reactions during or within 1 hour of infusion of irinotecan were treated with atropine at a dose of 1 mg intravenously by the subcutaneous route. Treatment cycles were repeated every 3 weeks. Radiographic studies were performed every 2 cycles of therapy for assessment of response. Efficacy was evaluated using Response Evaluation Criteria in Solid Tumors (RECIST).11 Toxicity was graded using the National Cancer Institute Common Terminology Criteria (version 2.0). Treatment cycles were continued until disease progression, unacceptable adverse events, or withdrawal of informed consent. The use of prophylactic granulocyte–colony–stimulating factors was not allowed.

Dose Modifications for Toxicity

Dose modifications were permitted for patients who experienced toxicity. Before each day of chemotherapy administration, the patients were required to have an ANC ≥1500/μL, a platelet count ≥100,000/μL, and peripheral neuropathy and diarrhea of ≤grade 1 in severity. For patients who did not meet the criteria on Day 1 of each cycle, the dose was delayed by 1 week. When the toxicity resolved to ≤grade 1 in severity, subsequent doses were administered without dose modifications. For patients who did not meet the criteria on Day 8 of each cycle, the planned dose was suspended.

In the event of grade 4 toxicity during a cycle, the next dose of the causative drug was reduced by 1 dose level. For purposes of dose reduction, the dose of paclitaxel and irinotecan were reduced by decrements of 15 mg/m2 and 10 mg/m2, respectively. A maximum of 2 dose reductions were allowed per patient. The dose of irinotecan was reduced for diarrhea of grade ≥2. The dose of paclitaxel was reduced in the event of neuropathy of ≥grade 2. The dose of both agents was reduced in the event of grade 4 neutropenia for >7 days, grade 3 or 4 neutropenia associated with fever, or grade 4 thrombocytopenia or anemia. Dose re-escalation was not allowed.

Patient Evaluations

Baseline evaluations were conducted within 1 week before the initiation of protocol therapy. A history and physical examination, complete blood count with differential, serum chemistry, vital signs, serum pregnancy test for women of childbearing potential, and assessment of PS were required at baseline. Radiographic studies including computerized axial tomography scan were performed within 4 weeks of study entry and every 2 cycles thereafter. Toxicity was assessed on each day of chemotherapy and at the initiation of every cycle. Bone scans and magnetic resonance imaging scans were only required when prompted by appropriate symptoms. Routine imaging of the brain was not necessary in asymptomatic patients.

Statistical Analysis

The primary endpoint of the study was response rate. The Simon 2–stage design was used. A response rate of 30% with the experimental regimen was considered to be of interest and a response rate <15% was classified as not worthy of further study. The estimated sample size was 55 patients. A total of 23 patients were to be accrued to the first stage of the study. If ≥4 responses were observed in the first stage, then the study would proceed to full accrual of an additional 32 patients. This design had a type I error of 10% with a power of 90%. Secondary endpoints included assessment of toxicity, time to disease progression, and overall survival.


Patient Baseline Characteristics

A total of 55 patients were entered into the study between the years 2004 and 2007 (Table 1). Four patients never initiated protocol therapy after registration. The reasons included the decision to seek other therapy (1 patient), progression of symptoms that required radiotherapy before the initiation of protocol therapy (1 patient), and logistical issues (2 patients). Fifty-one patients received at least 1 dose of protocol therapy and were considered evaluable for both response and toxicity. Twenty-three (45%) of the patients were women. The median age was 61 years. A total of 23 patients had an ECOG PS of 0, and 26 patients had a PS of 1. All patients had received prior chemotherapy, and 29 had received prior radiotherapy.

Table 1. Patient Baseline Characteristics
  1. ECOG indicates Eastern Cooperative Oncology Group.

Total no. of patients55
No. of patients evaluable51
Median age (range), y61 (33-86)
No. of females23 (45%)
ECOG performance status 
Prior therapy 
Response to prior chemotherapy 

Treatment Delivery and Toxicity

The median number of treatment cycles administered was 3. Fifteen patients received ≥6 cycles of treatment (Table 2). Disease progression was the most common reason for discontinuation from study. Three patients discontinued treatment due to toxicity (fatigue and diarrhea). Two patients decided to come off study due to logistical reasons including difficulty in commuting to the treatment center. There were no treatment-related deaths reported. Overall, the treatment was tolerated well. The majority of the patients completed protocol therapy without the need for dose modifications. Toxicities of ≥grade 3 were noted in 17 patients, of which 10 occurred during the first cycle of therapy. The majority of the toxicities reported in the first cycle were of grade 1 or 2 severity (Table 3). Reported grade 3 toxicities included allergic reactions, bone marrow suppression, fatigue, diarrhea, and laboratory abnormalities. Grade 3 or higher neutropenia was noted in 8 treatment cycles. Five episodes of grade 3 fatigue were recorded. There was no neuropathy of ≥grade 3 reported. Hyponatremia of ≥grade 3 was noted in 2 patients.

Table 2. Maximum Number of Cycles Administered Per Patient
No. of CyclesNo. of Patients (%)
110 (19)
35 (10)
48 (15)
52 (4)
611 (22)
71 (2)
91 (2)
102 (4)
Total51 (100)
Table 3. Adverse Events and Number of Patients by Worst Grade Adverse Event (All Cycles)
CategoryType of Adverse EventGradeaTotal
  • WBC indicates white blood cell; ANC, absolute neutrophil count; AGC, absolute granulocyte count.

  • a

    Toxicity was graded using the National Cancer Institute Common Terminology Criteria (version 2.0).

Allergy/immunologyAllergic reaction/hypersensitivity (including drug fever)001001
Blood/bone marrowHemoglobin101810029
 Leukocytes (total WBC)91560030
 Neutrophils/granulocytes (ANC/AGC)51062023
Cardiovascular (general)Hypotension010001
Constitutional symptomsFatigue (lethargy, malaise, asthenia)19650030
 Rigors, chills100001
 Sweating (diaphoresis)100001
 Urticaria (hives, welts, wheals)100001
 Diarrhea patients without colostomy14730024
 Stomatitis/pharyngitis (Oral/pharyngeal mucositis)800008
Infection/febrile neutropeniaInfection without neutropenia100001
MusculoskeletalMuscle weakness (not due to neuropathy)010001
 Neuropathic pain100001
PainAbdominal pain or cramping7210010


All 51 patients who received at least 1 dose of therapy were considered evaluable for efficacy. Seven patients were alive at the time of last follow–up. Although the study met its efficacy goals during the first stage of accrual, the overall response rate of 21% did not meet the anticipated response rate of 30%. Among 28 patients who were sensitive to first-line chemotherapy, there were 7 partial responses achieved, for an overall response rate of 25%. Among the 14 patients who had refractory disease to prior therapy, only 2 responses were noted, for an objective response rate of 14%. The outcome data with first-line chemotherapy was not available for 9 patients. The differences in response rate between patients with sensitive disease and those with refractory disease was not found to be statistically significant (P = .6925), in this post hoc analysis. In addition, 36% of the patients experienced disease stabilization. The overall median survival was 25.4 weeks (Fig. 1). The 6-month survival rate was 47%, and the 1-year survival rate was 22%. The median progression-free (PFS) survival for patients who had refractory disease to first-line therapy was 2 months compared with 3 months for patients who were sensitive to prior therapy. The median overall survival for the sensitive and refractory populations was 7.6 months and 5.5 months, respectively.

Figure 1.

Kaplan-Meier curve for overall survival. Pr indicates probability.


Patients with progression of SCLC after platinum-based chemotherapy have a poor overall prognosis. The initial approval of topotecan for patients with recurrent sensitive SCLC was based only on qualitative benefits. When single-agent topotecan was compared with placebo in a randomized phase 3 trial for patients with recurrent SCLC,12 the median survival was 26 weeks with topotecan compared with 13 weeks with best supportive care. There were improvements in several cancer-related symptoms in patients treated with topotecan compared with supportive care. Topotecan has also proven efficacious in combination with cisplatin in patients with previously untreated SCLC.13 Although topotecan has documented activity in both chemotherapy–naive patients and patients with recurrent SCLC, the evaluation of novel agents and regimens to improve the outcome of patients with SCLC is urgently needed.

The current study evaluated a combination regimen of paclitaxel and irinotecan for recurrent SCLC. To our knowledge, this is the first study to evaluate this combination in the second-line treatment setting for SCLC. The majority of the patients included in the study had sensitive disease to prior chemotherapy. Despite excellent tolerance of the treatment regimen, the response rate was modest at 21%. The median PFS and overall survival results were also modest. For patients with sensitive disease, the median survival was 7 months. Despite the activity of this regimen in patients with refractory disease, it is unlikely to represent a major step forward in the treatment of recurrent SCLC. Because the study did not meet the target to reject null hypothesis (response rate of ≤30%), no further follow-up is planned.

Irinotecan has recently garnered a great deal of interest as a treatment option for SCLC, based on a phase 3 study by Noda et al.14 Their study randomized patients to treatment with cisplatin in combination with either irinotecan or etoposide. The study was closed early due to the superior efficacy noted with the cisplatin-irinotecan regimen and led to the adoption of this regimen as a new standard in the Japanese patient population. However, 2 randomized phase 3 studies conducted outside Japan have failed to confirm the superiority of the cisplatin-irinotecan regimen in the first-line treatment of patients with extensive–stage SCLC.15, 16 The higher sensitivity to irinotecan noted in the Japanese patient population may be related to unique pharmacogenomic characteristics and warrants further evaluation. In the United States, the cisplatin-irinotecan regimen is no longer considered an evidence-based option for the first-line therapy of SCLC.

Overall, to our knowledge, very little progress has been made in the treatment of SCLC in recent years. Amrubicin, an anthracycline, has demonstrated promising activity in phase 2 studies and is currently under phase 3 investigation for patients with refractory or recurrent SCLC.17 In addition, several molecularly targeted agents are currently under investigation. However, the initial results reported with some of the targeted agents have been disappointing. Imatinib, an inhibitor of the c-kit receptor, failed to demonstrate any anticancer activity in patients with SCLC.18 Cediranib, an orally administered, potent vascular growth factor receptor inhibitor was recently tested as second-line therapy for patients with recurrent or refractory SCLC.19 No objective responses were noted, and the study did not proceed to the second stage of its accrual design. Agents that target the Bcl-2 pathway have also been under evaluation during recent times. Oblimersen, an antisense oligonucleotide against Bcl-2, was evaluated in combination with standard chemotherapy in a phase 2 study of extensive–stage SCLC.20 The results were indicative of an inferior hazards ratio for both PFS and overall survival with the addition of oblimersen. Recently, agents that inhibit various Bcl-2 family proteins have entered clinical trials in SCLC based on promising preclinical data.21 Although the results presented to date with targeted agents in SCLC have been disappointing, it is hoped that an improved understanding of the unique biologic characteristics of SCLC will lead to the development of newer classes of agents to combat this disease.


Supported in part by Pfizer Pharmaceuticals.