We compared concurrent combination chemotherapy and radiotherapy with surgery and adjuvant radiotherapy in patients with stage III/IV nonmetastatic squamous cell head and neck cancer. Patients with non-nasopharyngeal and nonsalivary resectable squamous cell head and neck cancer were randomised to receive either surgery followed by adjuvant radiotherapy (60 Gy over 30 fractions) or concurrent combination chemotherapy and radiotherapy (66 Gy in 33 fractions). Combination chemotherapy comprised two cycles of i.v. cisplatin 20 mg m À2 day À1 and i.v. 5-fluorouracil 1000 mg m À2 day À1 , both to run over 96 h given on days 1 and 28 of the radiotherapy. A total of 119 patients were randomised. At a median follow-up of 6 years, there was no significant difference in the 3-year disease-free survival rate between the surgery and concurrent chemoradiotherapy (50 vs 40% respectively). The overall organ preservation rate or avoidance of surgery to primary site was 45%. Those with laryngeal/hypopharyngeal disease subsite had a higher organ-preservation rate than the rest (68 vs 30%). Combination chemotherapy and concurrent irradiation with salvage surgery was not superior to conventional surgery and postoperative radiotherapy for resectable advanced squamous cell head and neck cancer. However, this form of treatment schedule with a view to organ-preservation can be attempted especially for those with laryngeal/hypopharyngeal and possibly oropharyngeal disease subsites. The majority of the patients with squamous cell head and neck cancer (SCHNC) present with locally and/or regionally advanced disease and the use of radical surgery and/or radiotherapy in this setting yield low locoregional control rates and 5-year survival rates not exceeding 40%.The administration of chemotherapy and radiotherapy concurrently makes use of the resultant synergistic activity to improve tumour cell kill. This strategy has found success in anal canal carcinoma, allowing high cure rates while obviating the need for radical surgery. Studies carried out in the 1990s using combination chemotherapy with concurrent radiation in SCHNC have shown this treatment approach to be feasible despite the significantly higher toxicity and have produced encouraging results. Adelstein et al in a phase II trial using cisplatin (CDDP) and 5-fluorouracil (5FU) combination with concurrent split-course radiotherapy have reported a 4-year relapse-free survival of 45% and an overall survival of 49% (Adelstein et al, 1993). This when compared retrospectively with a similar patient population treated with radiation alone was shown to be improved. Adelstein next investigated the use of the same combination regimen concurrently with a continuous course of radiotherapy (Adelstein et al, 1994). In 19 patients treated in this fashion, despite significant toxicity, there were no treatment-related deaths. At a median follow-up of 20 months, the projected Kaplan -Meier estimate of locoregional disease control was 92%, with the projected relapse-free survival of 86%. Of significance...

Patients with metastatic nasopharyngeal carcinoma have variable survival outcomes. We previously designed a scoring system to better prognosticate these patients. Here, we report results on validation of this new prognostic index score in a separate cohort of patients. Clinical features and laboratory parameters were examined in 172 patients with univariate and multivariate analyses and a numerical score was derived for each independent prognostic variable. Significant independent prognostic variables and their scores assigned included poor performance status (score 5), haemoglobin o12 g dl À1 (score 4) and disease-free interval (DFI) (DFIp6 months (score 10) or metastases at initial diagnosis (score 1)). Maximum score was 19 and patients stratified into three prognostic groups: good, 0 -3; intermediate, 4 -8; poor, X9. When applied to a separate cohort of 120 patients, 59 patients were good, 43 intermediate and 18 poor prognosis, with median survivals of 19.6 (95% CI 16.1, 23.1), 14.3 (95% CI 12.3, 16.2) and 7.9 (95% CI 6.6, 9.2) months, respectively. (logrank test: P ¼ 0.003). We have validated a new prognostic score with factors readily available in the clinics. This simple score will prove useful as a method to prognosticate and stratify patients as well as to promote consistent reporting among clinical trials.

A practically useful measure of quality of life should be simple and quick to complete. A shortened Chinese version of the Functional Living Index -Cancer (FLIC) was recently proposed and was called Quick-FLIC. This study aims to assess the measurement properties of the Quick-FLIC. A total of 190 patients who received care from the National Cancer Centre of Singapore completed a questionnaire package at baseline. Patients filled in a retest questionnaire on average 2 weeks after baseline to assess test -retest reliability and responsiveness to change. The Quick-FLIC scores correlated well with the Functional Assessment of Chronic TherapyGeneral scores (r ¼ 0.78). Patients with different treatment status, performance status and self-rated health had significantly different Quick-FLIC scores in the expected directions (ANOVA; each Po0.001). Internal consistency (Cronbach's alpha ¼ 0.87) and 2-week test -retest reliability (intraclass correlation ¼ 0.81) were also satisfactory. The measure was responsive to changes in health status (Po0.001). The Quick-FLIC is a valid and reliable measure of health-related quality of life of cancer patients. The shortening of established health-related quality of life instruments should be considered in order to reduce the burden of having patients to answer lengthy questionnaires.

We retrospectively analysed the results of patients with advanced non-small-cell lung cancer treated with gefitinib to derive clinical factors predictive of response and a favourable survival outcome. Patients were treated with gefitinib 250 mg per day and reevaluated 4 -8 weeks later with repeat CT scan and every 8 weeks thereafter to assess response and the duration of response. Pathology review by a histopathologist was conducted, in particular to confirm a recently published result of bronchioloalveolar carcinoma histology or its components as predictive of response to gefitinib. Logistic regression and Cox regression analytical methods were applied to determine factors that could predict for response and improved overall survival. A total of 110 patients were treated. The overall response rate was 32% partial responses (PRs). Only never-smoking status was predictive of response in the logistic regression analysis, adjusted OR ¼ 6.1, 95% CI ¼ 1.7, 21.5. The presence of a PR and good performance status were predictive of a favourable survival outcome from the Cox regression modelling. Responders had an adjusted HR of 3.0, 95% CI ¼ 1.5 -5.8 compared to nonresponders, while patients with ECOG status 0 -1 had an adjusted HR of 0.42, 95% CI ¼ 0.25 -0.72, compared with patients with ECOG status 2 -4. Bronchioloalveolar carcinoma or its components were distinctly absent on pathology review. In conclusions, Never-smoking status is an important clinical predictor of a favourable response to gefitinib.

The Simon two-stage minimax design is a popular statistical design used in Phase II clinical trials. The analysis of the data arising from the design typically involves the use of frequentist statistics. This paper presents an alternative, Bayesian, approach to the design and analysis of Phase II clinical trials. In particular, we consider how a Bayesian approach could have affected the design, analysis and interpretation of two parallel Phase II trials of the National Cancer Centre Singapore, on the activity of gemcitabine in chemotherapy-naïve and in previously treated patients with metastatic nasopharyngeal carcinoma. We begin by explaining the Bayesian methodology and contrasting it with the frequentist approach. We then carry out a Bayesian analysis of the trial results. The conclusions drawn using the Bayesian approach were in general agreement with those obtained from the frequentist analysis. However they had the advantage of allowing for different and potentially more useful interpretations to be made regarding the trial results, as well as for the incorporation of external sources of information. In particular, using a Bayesian trial design, we were able to take into account the results of the parallel trial results when deciding whether to continue each trial beyond the interim stage.