Comparing docetaxel with gemcitabine as second-line chemotherapy in patients with advanced non-small cell lung cancer : A single institute randomized phase II study

Background: Platinum-based doublet chemotherapy is the backbone of treatment in advanced non-small cell lung cancer (NSCLC) however second-line treatment options are controversial particularly in patients with borderline performance status (PS) of 2. The aim of this study was to compare efficacy and toxicity of weekly docetaxel versus gemcitabine in this clinical setting. Patients and methods: A total of 70 patients with advanced (stage IIIB, IV) NSCLC entered this single institute study. Cases of this study had experienced disease progression after the first-line platinum-based doublet chemotherapy, with PS 02 in “Eastern Cooperative Oncology Group” scale. They were randomly assigned by stratified blocks to receive docetaxel 35 mg/m2 (Arm A, n34) or gemcitabine 1000 mg/m2 (Arm B, n36) days 1, 8 and 15 every three weeks, for up to six cycles. Primary end point was progression free survival (PFS) and secondary end points were objective response rate, disease control rate, median overall survival (OS) and toxicity. Dose modification was permitted upon clinician’s discretion for each individual patient. Results: Median of PFS was 2.03 months in arm A and 2.63 months in arm B (HR 1.279; 95% CI: 0.710-2.304, P 0.551). Although median OS for arm A was numerically greater (9.2 months) than arm B (8.3 months) it was statistically non-significant (HR 1.384; 95% CI: 0.632 to 2.809, P 0.59). Objective response was higher in Arm B than that in Arm A (P 0.20) but disease control rates were statistically different in both arms (P 0.034). Statistically significant differences in term of leukopenia was seen in arm B (P 0.013). Conclusion: This study, with limited number of cases, indicates that in advanced NSCLC, weekly docetaxel and gemcitabine are reasonable second-line treatment options with statistically similar effectiveness in terms of PFS and median OS with manageable toxicities in patients with PS 0-2.

Nearly all patients who receive first-line chemotherapy finally experience relapse of the disease.Many of them have a good performance status and are suitable candidates for second-line chemotherapy.Choosing optimal chemotherapy protocol depends on multiple factors such as tolerability profile, histologic subtype, cost and patients preference.There are several options for secondline systemic therapy but among them, monotherapy with docetaxel [11,12], pemetrexed [13,14], (in nonsquamous NSCLC) and erlotinib [15] have been approved in disease recurrence/progression following prior chemotherapy.Gemcitabine has also been studied as second-line therapy but has not approved in this setting [16,17].This study was conducted to compare clinical benefits and toxicities of weekly docetaxel and gemcitabine in patients with recurrent/progressive advanced NSCLC who had been treated with platinum-doublet chemotherapy as the firstline therapy and have a borderline performance status.

Study design
This phase II, randomized, single blind, controlled, single institute study was performed to compare intravenous (IV) docetaxel with gemcitabine in patients with locally advanced or metastatic NSCLC previously exposed to a platinum-containing regimen.The primary end point was progression free survival (PFS) and secondary end points included response rate and overall survival (OS).Patients were allowed to enter the study 30 days after the prior chemotherapy with full recovery from all toxic effects.Informed written consent was obtained prior to participation in the study based on the ethics and scientific committees of Shahid Beheshti Medical University, and the study was conducted in compliance with the Helsinki declaration.

Eligibility criteria
Eligible patients had to meet the following criteria: aged over 18 years, with histologically confirmed NSCLC, initially at stage IIIB and IV (according to AJCC, 6 th edition) [18] with recurrence/progression after the previous doublet-platinum based chemotherapy regimen.They had a performance status (PS) of 2 in the Eastern Cooperative Oncology Group performance status scale.Other eligibility criteria were as follows: at least one unidimensionally measurable or assessable disease, adequate bone marrow reserve, serum creatinine less than or equal to 1.5 mg/dL or a calculated creatinine clearance greater than or equal to 60 mL/min, bilirubin level less than or equal to 2.0 mg/dL, AST less than or equal to twice the institutional upper limits of normal, or less than or equal to four times the institutional upper limits of normal if the patient had liver metastasis.Patients with significant or uncontrolled cardiac, metabolic or infectious diseases as well as patient with symptomatic brain metastasis were excluded.Previous treatment with docetaxel or gemcitabine was also permitted.

Treatment plan
Patients were randomly assigned in a 1:1 ratio to receive docetaxel 35 mg/m 2 (Arm A, n34) in an hour infusion or gemcitabine 1000 mg/m 2 (Arm B, n36) in 100 ml 5% dextrose which was administered over 30 min. in on days 1, 8 and 15 of a 28-day cycle.Dexamethasone (8 mg) twice daily was given the day before and after docetaxel infusion and also 30 min.before chemotherapy.In the absence of progressive disease or intolerable toxicity, the patients were treated for a minimum of four cycles.Patients who achieved a complete or partial response (CR or PR) could receive two additional cycles of therapy, for a maximum of 6 cycles.Presence of progressive disease, intolerable and serious side effect or patients refusal would lead to treatment discontinuation.Chemotherapy with docetaxel was discontinued if the patient experienced a significant hypersensitivity reaction or unacceptable toxicity (e.g., fluid retention, neurologic side effects of WHO grade 3 to 4, severe skin reactions, or serious organ toxicity).
All patients had a complete blood cell (CBC) count and biochemistry study, at baseline and prior to each therapy.Notably, in this real world study the dosage of cytotoxic agents was permitted to be adjusted by clinicians discretion based on adverse events in the course of treatment for each individual.Dose modification and concomitant granulocyte-colony stimulating factor (G-CSF) were allowed during the course of treatment with respect to the grade of neutropenia.The dose of cytotoxic agents was attenuated by 25% if patients experienced neutropenia at 1,000-1,500/dL and/or platelets 75,000-100,000/dL.If neutrophil or platelet count were less than 1,000/dL and 75,000/dL, respectively chemotherapy was postponed.Chemotherapy was delayed for a maximum of two weeks until recovery if the granulocyte count was  1,000 /l or the platelet count was  50,000/l.Gemcitabine administration would be delayed if bilirubin levels were 2 mg/dL.

Toxicity and response evaluation
After the initial two cycles, response was assessed according to "Response Evaluation Criteria in Solid Tumors" (RECIST) criteria [19].Objective response was comprised of CR  PR.Combination of CR, PR and stable disease (SD) was defined as "disease control" rate.Patients were evaluated after every cycle for any response based on physical examination and chest X-ray.Chest CT scan was requested after every 2 cycles and/or at the termination of protocol.After completion of therapy, patients were monitored at 8 weekly intervals until disease progression or death.The OS was calculated for all patients who died, from the date of registration to the date of death.Otherwise, the patient who was confirmed to be alive, was censored until the last day.PFS was calculated from the date of registration to the date of progression or death.
All patients who received at least one cycle were assessed for safety, toxicity assessment was based on "common terminology criteria for adverse events (CTCAE)" version 3.0 [20] criteria for withdrawal from study were unacceptable toxicity as determined by the treating physician in consultation with the study coordinator, a delay in treatment greater than 2 weeks, requirement for palliative radiotherapy, or patient refusal.

Statistical analysis
This study was designed to enroll a total of 70 patients.The sample size was determined based on a significance level of 5% for alpha to test the hypothesis if two treatment arms were equal in term of response rate, PFS and toxicity.For testing the differences in categorical variables between the two arms, the Chi-square test or Fisher's exact test was used.The difference in quantitative variables of both groups was compared by the Student's t-test or non-parametric Mann-Whitney test.Kaplan Meier's survival curves were obtained and the log-rank test was used to assess the significance of differences for PFS and OS between the two study groups.COX-PH regression model was used to estimate hazard ratios and their 95% CIs (confidence intervals).The analysis was "intention to treat" and includes` all randomized patient regardless of subsequent withdrawal from treatment or deviation from the protocol.All analysis was performed by SPSS version 16.

Results
From August 2007 to January 2010, a total of 70 eligible patients were randomly assigned in one of the two arms (Figure 1).There was no deviation from assigned treatment in our trial and also, no exclusion occurred after randomisation.The age range of patients was between 24 and 85 years with the mean of 55.50  13.46.Patient's characteristics in both arms at baseline are shown in Table 1.There was no significant difference in

Response and survival
Although median overall survival for docetaxel arm was numerically longer (9.2 months) than that for gemcitabine arm (8.3 months) but it was not statistically significant (P= 0.59) (Figures 2 and 3).Objective response rate was non-statistically higher in gemcitabine arm (arm A  5.8%, arm B 16.6%, P0.20   Abbreviation: CR = complete response; PR = partial response; SD = stable disease; Prog = progressive disease; NA = not available.

Toxicity
All patients in both arms were assessable for toxicity.
Important toxicities have been outlined in Table 3. Significant leukopenia was seen in Arm B (P0.013).
Neuropathy was more common in docetaxel arm.In contrast, nausea and vomiting, as well as neutropenia were more frequent in gemcitabine arm but not statistically significant (Table 3).No patient developed febrile neutropenia or significant renal toxicity in this study.No death related to drug toxicity was reported.

Discussion
This study was conducted to evaluate the effectiveness of weekly docetaxel versus gemcitabine in real world practice as the second-line therapy.The present study demonstrated that significant higher disease control rate and possibility of achieving numerically (but nonsignificant) objective response rate with gemcitabine second-line chemotherapy and considerable longevity of median OS in docetaxel arm.Both regimens had very manageable toxicity with no difference in the incidence of major side effects.This is when both arms are relatively identical in terms of patients' age, gender ratio, and histologic subtypes.Second-line systemic therapies might improve survival of patients with advanced NSCLC with recurrent/progressive disease after the first-line treatment with doublet platinum based chemotherapy compared with best supportive care [21,22].
There are several agents available for second-line treatment and in standard practice; clinicians are required to carefully choose the optimal option with respect to histologic subtype of tumor, EGFR and ALK status, as well as patient's performance status and preference.
The "National Comprehensive Cancer Network" (NCCN) has considered docetaxel, pemetrexed and erlotinib as standard second-line treatment options [23].However, several attempts in the world aim to improve the outcome in clinical trials with combination therapy or maintenance therapy (continuation or switch) [24][25][26][27].
Unfortunately, most patients at the time of disease recurrence/progression after first-line combination chemotherapy are minimally frail.Thus, monotherapy with a cytotoxic agent with weekly schedule is a much tolerable regimen for most of them in comparison to 3-weekly regimens.Docetaxel with every three-week schedule causes significant toxicity such as grade 3 to 4 neutropenia in 40% to 60% of patients and its side effects may deteriorate the physical well-being of patients [12].Weekly docetaxel has been experienced in phase II studies in breast [28], prostate [29] and NSCLC [30] and its related toxicity is more manageable [31].The response rate to weekly docetaxel in second and higher lines of therapy in advanced NSCLC is 10%-24 % [32,33].Result of this study with respect to response rate is similar to Shepherd et al. [11] that reported 6% partial responses.
Single agent gemcitabine, as second-line therapy in advanced NSCLC is an effective therapy in terms of symptom control and improvement of quality of life in comparison to best supportive care alone [34][35][36].However, this cytotoxic agent has received no approval from Food and Drug Administration (FDA) for this setting.
Response rate in Arm B is similar to Crinò et al. [16] study with about 16% response rate.
In this study, primary end point was PFS because few studies report significant differences in OS with chemotherapy in advanced NSCLC.Additionally, PFS is not influenced by other lines of therapy.Also, in many studies and in our study PFS was prolonged but it was not statistically significant.It should be considered that statistical significance in relation to clinical significance needs more attention and sample size plays an important role to reach clinically significant result [37].PFS results in our study were partially similar to the other studies with single agent second-line chemotherapy with docetaxel and gemcitabine [11,38].Our patients showed less sever toxicity in both arm compared with the other studies [11,17,39].
The importance of this study is more prominent in developing countries, where potential toxicities of therapies and their managements are major concerns of clinicians who treat patients with advanced NSCLC with palliative intent.In some parts of these countries specialized centers as well as expertise medical staff and physicians are not available to deal with major chemotherapy toxicities.For this reason, two weekly regimens were chosen for this study which has more manageable toxicities.Combination chemotherapy in second-line setting is associated with significant increase in some toxicity [24,40].Besides, the clinicians in this study were permitted to modify dosages according to patients' age, frailty, comorbid conditions, or even their clinical judgment.Moreover, in the most of the developing countries newer cytotoxic agents such as pemetrexed or molecular targeted agents such as erlotinib are not affordable for most of the patients or not fully funded by public health care system as a second-line or maintenance therapy.Bearing all of these in mind, in patients with advanced NSCLC who experience disease progression after platinum-doublet chemotherapy, weekly docetaxel or gemcitabine are the two treatment options in clinical practice, although they are not necessarily optimal therapies in some cases.
A relevant limitation of this study is the difference in firstline treatment schedules among both arms which could have masked the impact of second-line chemotherapy on OS or toxicity.

Conclusion
This study demonstrated that two regimens of weekly docetaxel and gemcitabine are comparable in terms of PFS with manageable toxicities.We propose that effectiveness of different standard chemotherapy regimens (along with biologic agents) be evaluated in the real world setting in developing countries.The results of these studies might be different from the results reported from highly selected patient populations in phase II or III clinical trials, conducted in developed countries.The importance of these issues will be more evident when we consider diversity of treatment facilities, trained staff, financial constrains or even patients socio-cultural level as confounding factors that could have an impact on the selection of systemic treatment in patients with relapsed/ progressed NSCLC.

Figure 1
Figure 1 Diagram of the study and clinical trial flow chart.A total of 70 patients received study treatment consisting of at least one dose of docetaxel (Arm A; n = 34) or gemcitabine (Arm B; n = 36).

Figure 3
Figure 3 Overall survival (OS) in docetaxel Vs gemcitabine arm.Kaplan-Meier survival curve for overall survival (OS) in docetaxel vs gemcitabine arms.Non-significant differences was observed between two groups, (P=0.59).

Figure 2 7 Arm
Figure 2 Progression free survival (PFS) in intention to treatment population docetaxel Vs gemcitabine.Kaplan-Meier survival curve from onset of recurrence for the effect of single agent second-line chemotherapy on progression free survival.Non-significant shortened PFS was observed between two groups, (P=0.551).

Table 1
Demographic characteristics of patients in both study arms.
any of the characteristics listed between the two groups including age, gender, stage and histologic subtype.Mean administered cycles in arm A was 2.15 (1-6) versus 2.62 (1-6) in arm B (P0.146).Dose modification during the course of treatment was more common in gemcitabine arm compared to that in docetaxel arm (13 vs 1 patients, P0.003).Three patients in arm A and 4 patients in arm B were treated with chest radiation therapy prior to second-line therapy.

Table 2
Response to treatment in two study arms.

Table 3
Major toxicity profile in two study arms.
** Toxicity assessment was based on "Common Terminology Criteria for Adverse Events" (CTCAE) version 3.0.