Results of the IFCT 0002 phase III study comparing a preoperative and a perioperative chemotherapy (CT) with two different CT regimens in resectable non-small cell lung cancer (NSCLC)

7530

Background: Association of surgery and CT is standard for early-stage NSCLC. Meta-analyses showed comparable efficacy of adjuvant and neoadjuvant CT. The primary objective was to compare survival between two different CT strategies: all before surgery (PRE) versus perioperative (PERI). Methods: Between 2001 and 2005, 528 patients with a stage IA-II resectable NSCLC were randomized to 4 parallel arms (A: 2 GP + 2 GP in responders, then surgery, B:2 GP - surgery + 2 GP in responders, C: 2 TC + 2 TC in responders then surgery, D: 2 TC - surgery + 2 TC in responders; GP: Gemcitabine 1250 mg/m2/d1, 8 and cisplatin 75 mg/m2/d1 q3 wk; TC: Paclitaxel 200 mg/m2/d1 and carboplatin AUC 6, q3 wk). Results: 501 patients were operated on, 96.2% in the preoperative CT arms (PRE: A+C) and 95.8% in the perioperative CT arms (PERI: B+D). Ninety- day postoperative mortality was 4.9% and 4.2%, respectively. Pathological complete response was not significantly influenced by the number of preoperative cycles (PRE:8.6%, PERI:6.4%). In an intent-to-treat analysis, 3-yr survival was 67.8% and 68.6%, respectively (p=0.96). In responders, despite a dramatic difference in CT compliance (90.4% and 75.2% having received the 4 cycles, respectively, p=0.001), 3-yr survival was 75.1% and 79.5%, respectively (p=0.82). Survival did not differ with the CT regimen (GP versus TC, p=0.84). Three-yr survival increased from 68.1% in the PRE arms to 77.2% in the PERI arms in squamous cell carcinomas (SCC), and decreased from 67.7% to 61.6% in non SCC, respectively (Cox model interaction, p=0.35). Three-yr survival was 74.6% in the GP arms and 70.7% in the TC arms, in SCC, and was 64.2% and 65.4%, in non SCC, respectively (interaction, p=0.51). There was no interaction between CT strategy and stage. In stage II patients, 3-yr survival was 59.1% but 76.5% in responders, comparable to that of all stage I patients (72.9%). Conclusions: Despite an increased compliance of the all preoperative chemotherapy strategy, no difference was observed between the PRE and PERI arms. There might be an advantage for perioperative CT and for gemcitabine-based in SCC and for preoperative CT and for taxane-based in non SCC.

No significant financial relationships to disclose.

Evaluation of class III beta-tubulin (bTubIII) expression as a prognostic marker in patients with resectable non-small cell lung cancer (NSCLC) treated by perioperative chemotherapy (CT) in the phase III trial IFCT-0002

7526

Background: IFCT-0002 phase III trial compared two timings of CT in early lung cancer, all before surgery (PRE) versus PERIoperative, and two CT regimens, CDDP-Gem vs. CBDCA-Pac. 528 patients were randomized. Paraffin embedded post- chemo pathological specimens were collected in the 490 non complete responder patients for tissue expression of the putative biomarker beta-tubulin III (bTubIII). Methods: 423 surgical pathological specimens with enough remaining viable tumor tissue after neoadjuvant chemo were processed for immunohistochemistry as published in the Bio-IALT study. A semi-quantitative score was attributed taking account the number of stained cells and the intensity of staining. Semi-quantitative scores were studied as continuous variables, without any pre- determined cut-off. Multivariate analysis for progression-free (PFS) and overall survival (OS) were corrected with Bonferroni-Holm method for multiple analyses. Median follow-up was 42 months. Results: bTubIII was the only IHC marker significantly associated with poor PFS in univariate (p=0.014) or multivariate analysis, adjusted for histology, T and stage (HR= 1.50 [1.07–2.10]; p=0.020). In patients with a pathological specimen showing a bTubIII positive immunostaining, median PFS was 30.6 months, versus 60.1 months (HR=1.46 [1.08–1.99]) for bTubIII negative patients. bTubIII IHC score remained predictive of poor OS in univariate (p= 0.0065) as in multivariate analysis (p=0.019 with Bonferroni correction, HR=1.75 [1.15–2.68] ). Median OS was not reached for bTubIII negative patients whereas it was 71.7 months in patients with bTubIII immunostaining of any intensity score (HR=1.61, [1.11–2.35]). Conclusions: This study showed a dramatic negative prognostic impact for bTubIII immunostaining in resectable early lung cancer. A subset of bTubIII expressing patients with poor prognosis did not take any advantage from perioperative chemo. Hence, those patients could rather have beneficiated from personalized adjuvant treatment with alternative approaches.

No significant financial relationships to disclose.

[Preoperative chemotherapy in non-small cell lung cancer: advantages, disadvantages, level of evidence]

The combination of chemotherapy and surgery is a standard of care for non-small cell lung cancer, as shown by the recently published "Standards, Options et Recommandations" (SOR) by the Fédération des Centres de Lutte contre le Cancer. This document was approved by the INCa, the SPLF, the Ligue contre le Cancer, the IFCT. However, the respective position of chemotherapy and surgery remains debated. Most trials of preoperative chemotherapy were closed when the positive studies of adjuvant chemotherapy were published. Therefore, the trials of preoperative chemotherapy lack strength to conclude on the validity of the concept. Confirmation will come from meta-analyses. Two meta-analyses based on published data have been yet published, and are positive.

[IFCT-0302 trial: randomised study comparing two follow-up schedules in completely resected non-small cell lung cancer]

BACKGROUND

The authorities advocate a minimalist attitude towards the follow-up of resected bronchial carcinoma (clinical examination and chest x-ray). A survey showed that 70% of French respiratory physicians have chosen to use the CT scanner and often endoscopy. The published data are equivocal and are often based on retrospective studies. Lung cancer is a good model for a study of post-operative surveillance. Recurrences often occur in easily observed areas, they may be detected while still asymptomatic and are sometimes potentially curable. Second primary tumours may develop at the same site.

METHODS

The Intergroupe Francophone de Cancerologie Thoracique (IFCT) has initiated a trial comparing simple follow-up (clinical examination, chest x-ray) with a more intensive follow-up (CT scan, fibreoptic bronchoscopy). The surveillance will take place every 6 months for 2 years and then annually until 5 years.

EXPECTED RESULTS

The main aim is to determine whether intensive follow-up improves patient survival. The opposite question is equally important. If an expensive and demanding follow-up does not affect the chances of cure these results will influence our practice.

IFCT0002 phase III study comparing a preoperative (PRE) and a perioperative (PERI) chemotherapy with two different CT regimens in resectable non-small cell lung cancer (NSCLC): Early results

7519

Background: Recent trials have shown a survival benefit of CT in resectable NSCLC. The primary objective was to define the best timing of CT (all before surgery versus perioperative). Another objective was to compare two regimens, gemcitabine-cisplatin (GP) and paclitaxel-carboplatin (TC). Methods: Between May 2001 and Dec 2005, 528 patients (pts) with a stage IA-II resectable NSCLC were randomized to 4 parallel arms: A: 2 GP + 2 GP in responders, then surgery, B: 2 GP - surgery + 2 GP in responders, C: 2 TC + 2 TC in responders then surgery, D: 2 TC - surgery + 2 TC in responders (GP: Gemcitabine 1250 mg/m2/d1, 8 and cisplatin 75 mg/m2/d1 q3 wk; TC: Paclitaxel 200 mg/m2/d1 and carboplatin AUC 6, q3 wk). Results: Pathological tumor volume and pathological complete response rate did not differ with the number of preoperative cycles. Proportions of pts receiving cycles 3 and 4 were higher when all CT was given before surgery. There were several significant differences in the main toxicities between GP and TC. Conclusions: 1- GP and TC were effective and safe. 2- Results of pathological response suggested that 2 cycles might be as effective as 4 cycles. 3- Dose intensity was higher when all chemotherapy was given before surgery compared to both before and after surgery.

No significant financial relationships to disclose.

[Table: see text]

[Early stages of non small cell lung cancer (I. II. IIIA). Role of preoperative treatments]

The first meta-analysis of preoperative chemotherapy in non-small cell lung cancer (NSCLC) was published by Berghmans et coll. It included the first six reported studies and, despite the small number of patients involved, concluded in favour of preoperative chemotherapy. These six trials are summarized here. There are three other trials, which were not included in this meta-analysis; the SWOG study which was presented at the ASCO meeting in 2005 and an Italian and a Spanish trial, the results of which are still awaited. The advantages of preoperative chemotherapy are discussed. There are two other trials, whose designs are very different. The objective of the study by Albain et coll. was to evaluate the role of surgery after induction chemo-radiation in stage IIIa NSCLC. The results seem to be encouraging in patients who can undergo a (bi)lobectomy. The objective of the second trial was to evaluate whether surgery could improve survival after chemotherapy compared to thoracic irradiation in unresectable stage III disease. Although chemotherapy probably increased resectability, survival was not improved in operated patients.

[Treatment of localised lung cancer]

This paper focuses on stage I, II and IIIA non-small cell lung cancer treatable with local treatment. It addresses five questions raised by strategies combining local treatments with chemotherapy. Even if chemotherapy increases resectability of stage III disease, the chemotherapy-surgery combination has not been demonstrated to increase survival compared to the standard chemo-radiation treatment. The results of the study by Van Meerbeeck do not support this hypothesis. Does surgery, added to chemo-radiotherapy, improve the outcome in stage IIIAN2 disease? This was the question addressed by the study by K. Albain. There is probably not clear cut answer. However, the trimodality strategy might be interesting in patients undergoing a lobectomy and might have a negative impact when a pneumonectomy has been performed. In patients with a non resectable/inoperable cancer treated with standard chemoradiation, the concomitant strategy has been shown to be superior to sequential treatment. However, due to acute toxicity, it should be delivered to selected patients, who still need to be better defined. The chemotherapy-surgery combination is becoming standard (in stage II disease) and most cooperative groups will probably stand in favour of it in 2006. The best respective timing for chemotherapy and surgery is still debated. There are many advantages in favour of preoperative chemotherapy, including better feasibility and the higher proportion of patients who can benefit. However, there is no statistically reliable demonstration of such superiority.

A prospective randomized phase III, double-blind, placebo-controlled study of thalidomide in extended-disease (ED) SCLC patients after response to chemotherapy (CT): An intergroup study FNCLCC Cleo04 - IFCT 00–01

7057

Background: This study aimed at determining whether or not thalidomide prolongs survival of patients (pts) suffering from SCLC. Methods: Eligibility consisted of previously untreated ED-SCLC, age <70 years, PS ≤2, weight loss <10% and, for women, post-menopausal status. Pts were registered in the study and received two courses of PCDE given 4 weeks apart with G-CSF primary prophylaxis recommended. Afterwards, pts who experienced a response were randomized to receive four additional cycles of PCDE plus thalidomide, (400 mg daily) or placebo. The planned accrual was 200 randomised pts in order to detect a 20% survival improvement. Results: The study was shortened with final analysis performed taking into account 119 registered pts (low accrual). There were 4 toxic-deaths (3.3%). Tumour assessment performed after the first two CT courses demonstrated 11 complete responders and 86 partial responders (81.4% overall response rate). Among these pts, 92 were randomly assigned, 49 in the thalidomide group and 43 in placebo group. The 5 remaining pts were not randomised due to poor recovery from previous CT. Pre-study pts’ characteristics did not differ between the two groups. The planned six cycles of PCDE were delivered to an equal proportion of pts in both groups (75.5% versus 74.4%). Mean ± SD exposure duration to thalidomide was 4.5 months ± 2.7 and to placebo 5.1 ± 2.4 (NS). Reasons for withdrawal differed between the two groups with toxicity as main reason for thalidomide (55.3% versus 35%) and disease progression as main reason for placebo (43% versus 62%; p = 0.06). In Cox model of overall survival within the 9 months following randomisation, pts allocated to the thalidomide group had the longest survival (HR of death for pts in the thalidomide group: 0.48 [95% CI: 0.24–0.93]; p = 0.03; median survival from randomisation: 11.7 versus 8.7 months for thalidomide and placebo groups respectively); Conclusion: Thalidomide prolongs survival of pts with SCLC after response to CT. This study is a clue in favour of angiogenesis process as therapeutic window in SCLC therapy. Supported by the French League against Cancer.

No significant financial relationships to disclose.

Long-duration, weekly treatment with gemcitabine plus vinorelbine for non-small cell lung cancer: A multicenter phase II study

In this phase II study, gemcitabine and vinorelbine were combined at suboptimal doses for weekly administration in advanced non-small cell lung cancer (NSCLC). The primary objectives were to determine objective response rate (ORR) and time to progression (TTP). Secondary endpoints were safety and overall survival. Chemonaive patients with histologically or cytologically confirmed stage IIIB or IV NSCLC received vinorelbine (25 mg/m2) immediately followed by gemcitabine (800 mg/m2) once each week (on day 1) for 6 months without rest. From May 1998 to May 1999, 40 patients were enrolled (85% males; 70% stage IV) with a median age of 65.5. A total of 478 doses were administered, with a median of 9 per patient (range 2-72). The ORR was 27.5% (95% CI, 15.1-44.1%). The median TTP was 3.5 months (95% CI, 2.9-4.4 months). At a median follow-up of 6.5 months, the median survival was 11.6 months, and survival rates at 1 and 2 year(s) were 47.5% and 15.8%, respectively. The most common grade 3/4 hematologic toxicity was neutropenia, in 70% of patients, with febrile neutropenia in 28%. The most common grade 3/4 non-hematologic toxicity was transaminase elevation, in 22.5% of patients, which was transient and reversible. The other most prominent toxicities were, unexpectedly, pulmonary and cardiac toxicities. Based on these results, weekly, long-term administration of gemcitabine-vinorelbine appears to be an active regimen in NSCLC that warrants further investigation.

[Pain management in bone metastasis of pulmonary origin: new interventional and metabolic techniques]

Invasion of bone by a metastatic lesion is the most common cause of pain in cancer patients. Pain management in these patients is an important and difficult task. The pain is not always properly controlled by high doses of specific medication, radiation therapy or chemotherapy. When these therapies do not provide adequate pain relief, percutaneous vertebroplasty, cementoplasty, radiofrequency ablation and internal radiotherapy appear to be elegant and efficient complementary alternative pain control methods.

[Pre-operative chemotherapy in non-small cell lung cancer]

Pre-operative chemotherapy for non-small cell bronchial carcinoma (NSCLC) has the twin objectives of destruction of micrometastases and increased resectabilty of the primary tumour. The trials of Rosell and Roth showed weak benefits in favour of the combination. In the French trial overall survival was no different for the whole group but was improved in early stage disease. Preliminary results of the SWOG trial show a non-significant difference in 2 year survival of 6%. The EORTC 08941 trial showed no difference between surgery and radiotherapy following induction chemotherapy in non-resectable stage IIIAN2 NSCLC. The INT-0139 trial compared surgery following induction chemo-radiotherapy with chemo-radiotherapy alone. There was no difference between the two strategies but analysis of sub-groups suggested that some groups might benefit from the triple combination. Two further trials await publication. The small number of patients in each trial suggests that a meta-analysis will be necessary to reach a definite conclusion. The combination of surgery and chemotherapy is becoming standard in stage II disease. Only the timing, pre- or post-operative, remains controversial. At present, of the original objectives, only the destruction of micro-metastases has been confirmed.

Gemcitabine–docetaxel versus cisplatin–vinorelbine in advanced or metastatic non-small-cell lung cancer: a phase III study addressing the case for cisplatin

BACKGROUND

This multicenter, randomized, phase III study compared the efficacy, including progression-free survival (PFS), and safety of gemcitabine-docetaxel (GD) combination versus cisplatin-vinorelbine (CV) in the treatment of advanced non-small-cell lung cancer (NSCLC).

PATIENTS AND METHODS

Chemonaive patients with stage IIIB or IV NSCLC were treated with GD (gemcitabine 1000 mg/m(2) days 1 and 8 plus docetaxel 85 mg/m(2) day 8, every 3 weeks for eight cycles) or CV (cisplatin 100 mg/m(2) day 1 plus vinorelbine 30 mg/m(2), days 1, 8, 15 and 22, every 4 weeks for six cycles).

RESULTS

A total of 311 patients were enrolled (155 GD and 156 CV). Neither PFS nor overall survival differed significantly between the two arms (median PFS 4.2 and 4 months; median survival 11.1 and 9.6 months; 1-year survival 46% and 42%, for GD and CV, respectively). For the GD arm compared with the CV arm, the hazard ratio for PFS was 1.04 [95% confidence interval (CI) 0.83-1.32], and for overall survival, it was 0.90 (95% CI 0.70-1.16). Objective response rates did not differ significantly (31% for GD, 35.9% for CV). Myelosupression, emesis and frequency of febrile neutropenia were less pronounced on the GD arm, whereas fluid retention and pulmonary events were more pronounced. The CV arm experienced a higher number of serious adverse events and a lower compliance with the protocol. There was no quality of life (QoL) difference between arms. Median time to definite impairment of health-related QoL was 153 and 168 days in GD and CV arms, respectively.

CONCLUSIONS

There was no advantage in PFS with GD compared with CV; however, the CV regimen had higher rate of toxic events, mainly myelosuppression. The herein, non-platinum-containing regimen could be considered as a rational alternative to the cisplatin-based doublet.

Multivariate analysis of factors predictive of brain metastases in localised non-small cell lung carcinoma

Brain metastases are a frequent feature of the course of non-small cell lung carcinoma (NSCLC). The potential usefulness of prophylactic cranial irradiation (PCI) has led to the search for target groups likely to derive benefit. This multivariate analysis looked for factors predictive of brain metastases in a group of stages I-III NSCLC patients under care of the thoracic oncology unit of Besançon University Hospital from 1977 to 2001. All the patients had the same follow-up. They were divided into two groups: BM+ when they had a brain metastasis as the first site of progression, whether solitary or not, and BM(-) otherwise. Variables analysed were age, gender, performance status (0-1 versus 2-3), weight-loss stage T-status, N-status, pathological type, type of treatment, administration of chemotherapy, use of cisplatin and response to treatment. Three hundred and five patients were eligible and there were 77 patients (25.25%) in the BM+ group. Median time to onset of brain metastases was 12 months (1-163 months) and median survival from the diagnosis of brain metastases was 6 months (1-65 months). Factors predictive of brain progression were age < or =62 years (RR: 2.5, 95% CI: 1.33-4.76 and P = 0.004), T4 tumour status (RR: 3.75, 95% CI: 1.72-8.21 and P = 0.0009), N2-3 (RR: 2.61, 95% CI: 1.32-5.15 and P = 0.0057), and adenocarcinoma (RR: 3.39, 95% CI: 1.78-6.46 and P = 0.0002). No aspect of treatment plays a role in the frequency of this type of metastasis. These factors predictive of brain progression could serve as a basis for the selection of patients with the aim of sitting of studies on prophylactic cranial irradiation in NSCLC.

[Pre- and peri-surgical chemotherapy of stage I and II resectable non-small cell lung cancers]

Chemotherapy (CT) combined with surgery in non-small cell lung cancers has been studied for a number of years. It can be used prior to or following surgery (adjuvant). A long rather unfruitful period ended with the meta-analysis of the Non-Small Cell Cancer Collaborative Group, published in the British Medical Journal in 1995 that suggested an increase in survival of 5% at 5 years with the addition of adjuvant chemotherapy to surgery. Since this publication, arguments have accumulated in favour of this combination. Phase II studies have shown the feasibility of pre-surgical CT. A randomised trial in France showed a near 10% improvement in survival at 5 years, approaching statistical significance, and that this beneficial effect was further enhanced in the early stages of cancer. Excess post-surgical morbidity and mortality, even though non-significant, emphasizes the need for an effective but less toxic CT than the mitomycine-ifosfamide-cisplatin combination initially selected. In the field of adjuvant CT, the arguments in favour of the association have accumulated with the positive results of 3 studies, the IALT trial, the BR10 trial of the Canadian National Cancer Institute and the 9633 trial of the CALGB, with the latter two studies presented this year at the American Society of Clinical Oncology meeting. Four other pre-surgery CT trials are ongoing, but their results will not be available for several years. When choosing optimal timing among the various CT administration methods, before or after surgery, the arguments are in favour of pre-surgery CT: the possibility of assessing the chemosensitivity of the tumor, permitting the early withdrawal of treatment if it fails (presently in 40% of patients), the enhanced acceptability of CT by the patients, and the increase in resectability of the tumours. Conversely, however, one must note the greater difficulty for staging and the increase in post-surgical risks, basically in N2 patients, which will gradually lead to its replacement by the use of 3rd generation CT.

Randomised, multicentre phase II study assessing two doses of docetaxel (75 or 100 mg/m2) as second-line monotherapy fornon-small-cell lung cancer

BACKGROUND

The survival benefit associated with first-line chemotherapy in advanced lung cancer led to the need for second-line chemotherapy. Docetaxel (Taxotere) has proven efficacy in both settings. This study evaluated the safety and efficacy of two doses of docetaxel in patients with non-small-cell lung cancer who had failed first-line platinum-based chemotherapy.

PATIENTS AND METHODS

In total, 182 patients from 24 French centres were randomised and treated with either docetaxel 75 mg/m(2) (arm A) or 100 mg/m(2) (arm B) every 3 weeks. Baseline characteristics were well balanced, except more patients in arm A had metastatic disease (91.4% versus 78.7%) and therefore the median number of sites involved for arm A was three compared with two for arm B.

RESULTS

Median time to treatment failure was 1.34 months [95% confidence interval (CI) 1.28-1.64] for arm A and 1.64 months (95% CI 1.34-2.62) for arm B. Median overall survival was 4.7 months (95% CI 3.8-5.9) for arm A versus 6.7 months (95% CI 4.8-7.1) for arm B. According to a blinded expert panel, disease control was achieved in 35 (43.8%) patients in arm A and 39 (49.4%) patients in arm B. More patients in arm B experienced grade 3-4 neutropenia (B: 72.7% versus A: 44.0%), asthenia (B: 20.2% versus A: 10.8%) and infection (B: 6.7% versus A: 2.2%). Three treatment-related deaths were reported in each arm.

CONCLUSIONS

The optimal docetaxel dosage in this second-line setting is 75 mg/m(2), as it has a more favourable safety profile and on balance a similar efficacy to the 100 mg/m(2) dose.