The use of differential item functioning analyses to identify cultural differences in responses to the EORTC QLQ-C30

INTRODUCTION

The European Organisation for Research and Treatment of Cancer (EORTC) QLQ-C30 is a widely used health-related quality of life instrument. The main aim of this study is to investigate whether there are international differences in response to the questionnaire that can be explained by cultural factors.

METHODS

Analyses involved a database of 106 separate studies including data from over 28,000 respondents. Differential item functioning (DIF) analyses using logistic regression were conducted for each item of the EORTC QLQ-C30 with respect to cultural/geographic group. Results were qualitatively compared with previously reported DIF analyses by translation to explore whether the source of the DIF was more likely to be linguistic or cultural in nature.

RESULTS

Although most response patterns were similar, there were a number of international differences in how the questionnaire was answered. The largest variations were found in the results for Eastern Europe and East Asia. Results for the UK, the US and Australia tended to be similar. Many of the European results followed patterns that were more clearly explained when grouped by translation than when grouped by geographical region.

DISCUSSION

Our results suggest that, in general, the EORTC QLQ-C30 is suitable for use in a wide variety of countries and settings. Some response variations that have the potential to affect the results of international studies were identified, but it was not always clear whether the source of the variation was primarily linguistic or cultural.

Breast cancer patients on adjuvant chemotherapy report a wide range of problems not identified by health-care staff

BACKGROUND

Adjuvant chemotherapy for primary breast cancer is associated with significant side effects. The aims of this study were (1) to compare health-related quality of life (HRQL) in patients undergoing adjuvant chemotherapy to patients not on chemotherapy and (2) to compare these results against a survey investigating health-care professionals' knowledge of HRQL.

METHODS

Patients on adjuvant cyclophosphamide, methotrexate, fluoracil chemotherapy were compared to 'low-risk' patients not on chemotherapy ('control group'). A questionnaire including the EORTC QLQ-C30, the Hospital Anxiety and Depression Scale (HADS), and the DBCG 89 Questionnaire was administered six times during a 2-year period. Forty-six experienced health-care professionals were asked which quality-of-life issues they thought were affected by adjuvant chemotherapy.

RESULTS

After 2 years, 159 of 242 patients on chemotherapy and 148 of 199 patients in the control group were alive and recurrence-free and had completed all questionnaires. Worse HRQL during chemotherapy was seen, as had been previously suggested, for 23 of 30 variables. A number of the health-care professionals had not indicated patients to have these side effects. Several side effects persisted after the chemotherapy.

CONCLUSIONS

This study provides the most comprehensive description of HRQL in adjuvant therapy to date. The discrepancy between patients and doctors/nurses suggests that patients have been insufficiently informed about the impact of chemotherapy on quality of life. The results of this study provide a basis for information that can be given to patients, and indicate that the care offered to patients in chemotherapy should seek to prevent, identify, and alleviate a very broad range of problems.

The clinical significance of adaptation to changing health: a meta-analysis of response shift

AIMS

When individuals experience changes in their health states, they may alter their internal standards, values, or conceptualization of quality of life (QOL). Such 'response shifts' can affect or distort QOL outcome measurement, which is of particular concern when evaluating medical or psychosocial interventions. Although clinicians and researchers acknowledge the occurrence of response shifts, little is known about the magnitude and clinical significance of those effects. To fill this gap in knowledge about response shift phenomena, we performed a meta-analysis on published QOL articles on response shift.

METHODS

Extensive literature searches and multiple contacts with researchers yielded a collection of 494 articles for potential reviewing. We retained only published longitudinal studies that measured response shift, resulting in 26, of which 19 reported the requisite data for computing an effect size (ES). We calculated and compared the ESs for each study with regard to potential moderator variables: the QOL domains measured, disease group investigated, sample size, and response shift method used. We rated studies for quality to allow ES weighting.

RESULTS

When we examined ES absolute values, we found that ES magnitude was small, with the largest ESs detected for fatigue, followed by global QOL, physical role limitation, psychological well-being, and pain (mean absolute value(ES(weighted)) = 0.32, 0.30, 0.24, 0.12, and 0.08, respectively). ESs varied considerably in direction. Aggregating raw ES scores over all studies led to positive and negative values canceling each other out (mean directional ES(weighted) = 0.17, 0.02, -0.01, 0.06, and 0.02, respectively). We found little evidence of an effect for the moderator variables examined.

CONCLUSIONS

A definitive conclusion on the clinical significance of response shift cannot currently be drawn from existing studies. For a number of reasons, ES estimates were primarily based on then-test results, a method that is not without criticism, such as its susceptibility to recall bias. We recommend a standardized approach for reporting results of future response shift research to advance the field and to facilitate interpretation and comparisons across studies.

Determinants of the intracluster correlation coefficient in cluster randomized trials: the case of implementation research

The objective of this research was to identify determinants of the magnitude of intracluster correlation coefficients (ICCs) in cluster randomized trials from the field of implementation research. A survey of experts was conducted to generate a priori hypotheses of factors that might affect ICC size. Hypotheses were tested on empirical estimates of ICCs calculated from 21 implementation research datasets, mainly from the UK. Effects of setting (primary or secondary care), type of variable (process or outcome), type of measurement (objective or subjective), prevalence of outcome and size of cluster were tested. In total, 220 ICCs were available (range 0 to 0.415). Significant differences in ICC magnitude were found. The ICCs were significantly higher for process than for outcome variables, and for secondary care outcomes compared with primary care outcomes. The effects of prevalence and size were less clear cut. There was no evidence to suggest that type of measurement affected ICC size. In conclusion, accurate estimates of ICCs are essential for sample size calculations for cluster randomized trials of professional behaviour change interventions. This study demonstrates that ICCs are sensitive to a number of trial factors, particularly setting and outcome type. These factors must be considered when planning such cluster randomized trials.

Comparing translations of the EORTC QLQ-C30 using differential item functioning analyses

The European Organisation for Research and Treatment of Cancer (EORTC) QLQ-C30 is one of the most widely used quality of life instruments for cancer patients. The aim of this study was to assess whether there were linguistic differences in the way an international sample answered the EORTC QLQ-C30 questionnaire. Thirteen translations of the EORTC QLQ-C30, representing 22 countries, were investigated using a database of 27,891 respondents, incorporating 103 separate studies. Differential item functioning (DIF) analyses were conducted using logistic regression to identify items which, after controlling for subscale, were answered differently by language of administration. Both uniform and non-uniform DIF were assessed. Although most languages showed similar results to English, at least one instance of statistically significant DIF was identified for each translation, and a few of these differences were large. In some cases, the patterns were supported by the results of qualitative interviews with bilingual people. Although, overall, there appeared to be good linguistic equivalence for most of the EORTC QLQ-C30 items, several scales showed strongly discrepant results for some translations. Some of these effects are large enough to impact on the results of clinical trials. Based on our experience in this study, we suggest that validation of translations of health-related quality of life instruments should include exploration of DIF.

A review of the quality of life of adult patients treated with long-term parenteral nutrition

BACKGROUND AND AIMS

Some previous studies have assessed quality of life (QoL) in home parenteral nutrition (HPN) using generic instruments or non-validated questionnaires. A systematic search of electronic databases and relevant publications identified 50 publications. This paper reviews the QoL of patients receiving HPN and discusses the factors affecting QoL.

RESULTS

There is little available data about the QoL of HPN patients. Both HPN and the underlying disease may affect QoL, and an evaluation of QoL requires the separation of these two issues.

CONCLUSIONS

There is a need for a standardised, scientifically validated, treatment-specific instrument to measure QoL in this population. The use of a treatment-specific QoL questionnaire should become part of the routine clinical management of HPN patients.

Chemotherapy versus ovarian ablation as adjuvant therapy for breast cancer: impact on health-related quality of life in a randomized trial

BACKGROUND

Ovarian ablation is an effective adjuvant therapy for primary breast cancer but little is known about its quality of life impact relative to the more widely used adjuvant chemotherapy. This randomized study compared quality of life outcomes of adjuvant ovarian ablation versus cyclophosphamide, methotrexate, fluoracil (CMF) chemotherapy.

METHODS

The Danish Breast Cancer Cooperative Group (DBCG) trial 89-b randomized premenopausal patients with receptor-positive, primary breast cancer between nine cycles of CMF chemotherapy given every 3 weeks and ovarian ablation by oophorectomy. In total, 317 randomized patients were invited to take part in a longitudinal quality of life study with assessments at 1, 3, 5, 9, 15, and 24 months after randomization. The questionnaire included the EORTC QLQ-C30, the Hospital Anxiety and Depression Scale, and additional items assessing potential symptoms not included in the standard instruments.

RESULTS

After 2 years, 260 women were alive and recurrence-free, and 196 of these (75%) had completed all six questionnaires. Overall, patients in the chemotherapy group had more symptomatology at the first three assessments (i.e., during the 6 months treatment period), except for hot flushes/sweats. There were few differences between groups at later assessments. In chemotherapy patients, the likelihood of preserving ovarian function decreased steeply with increasing age. CMF chemotherapy and ovarian ablation have similar impact on recurrence and survival.

CONCLUSION

Chemotherapy had more negative impact on health-related quality of life but preserved ovarian function in some younger patients.

The development of the EORTC QLQ-C15-PAL: a shortened questionnaire for cancer patients in palliative care

This study aimed at developing a shortened version of the EORTC QLQ-C30, one of the most widely used health-related quality of life questionnaires in oncology, for palliative care research. The study included interviews with 41 patients and 66 health care professionals in palliative care to determine the appropriateness, relevance and importance of the various domains of the QLQ-C30. Item response theory methods were used to shorten scales. Patients and health care professionals rated pain, physical function, emotional function, fatigue, global health status/quality of life, nausea/vomiting, appetite, dyspnoea, constipation, and sleep as most important. Therefore, these scales/items were retained in the questionnaire. Four scales were shortened without reducing measurement precision. Important dimensions not covered by the questionnaire were identified. The resulting 15-item EORTC QLQ-C15-PAL is a 'core questionnaire' for palliative care. Depending on the research questions, it may be supplemented by additional items, modules or questionnaires.

Which mini-mental state exam items can be used to screen for delirium and cognitive impairment?

Cognitive impairment is common in palliative care patients, but it is frequently undetected. The clinical consequence is that psychiatric states such as delirium, which often present with cognitive impairment, are inadequately treated. A short and simple questionnaire for screening of cognitive impairment is required for these patients, in order to proceed with more advanced testing if necessary. In this study, we explored the results from two samples of patients (n=290 and n=217) who had completed the Mini-Mental State Examination (MMSE). Cases of cognitive impairment are considered indicated by an MMSE score of less than 24 of the total 30. We found that caseness could be fairly accurately screened by using four of the original 20 MMSE items, and that a six-item questionnaire further greatly improved the discrimination.

The Norwegian Doloplus-2, a tool for behavioural pain assessment: translation and pilot-validation in nursing home patients with cognitive impairment

BACKGROUND

Pain assessment is challenging in cognitively impaired (CI) patients due to inadequate self-report skills and observational ratings are an alternative. The Doloplus-2 is developed for pain assessment in the CI and rates somatic, psychomotor and psychosocial behaviours as indicators of pain.

AIMS

To translate the Doloplus-2 into Norwegian, to test the Doloplus-2 with regard to criterion validity and to obtain the administrators' evaluation of the clinical performance of the Doloplus-2.

METHODS

Nurses at three nursing homes, in collaboration with two research assistants, administered the Doloplus-2 to 59 patients with dementia. The results were compared against experienced clinicians' pain ratings. Regression analyses were performed to explore each different item's contribution to the total pain score. The administrators also completed a debriefing questionnaire.

RESULTS

The instrument was translated according to international guidelines. Regression analyses demonstrate that the Doloplus-2 score accounts for 62% (R2) of the expert score and that the four most informative items could explain 68% of the expert score. Analyses of the different Doloplus-2 items indicate that facial expressions explain most and social life least of the expert's pain ratings. The administrators reported that Doloplus-2 was helpful and easy to administer, but questioned the validity of the psychosocial domain.

CONCLUSIONS

The Norwegian Doloplus-2 demonstrates satisfactory criterion validity and clinical value in this pilot study. However, the content of the instrument needs a general re-evaluation, especially with regard to the psychosocial items.

Immediate or delayed radiotherapy in advanced non-small cell lung cancer (NSCLC)? Data from a prospective randomised study

BACKGROUND AND PURPOSE

To compare the course of symptoms and health-related quality-of-life (HRQOL) after immediate thoracic radiotherapy (TRT) between symptomatic (S) and non-symptomatic (NS) patients with advanced NSCLC.

PATIENTS AND METHODS

407 stage III/IV patients were initially treated with immediate TRT within a randomised phase III trial comparing different fractionation schedules. At inclusion, patients were prospectively stratified according to presence (S) or absence (NS) of tumour-related chest/airway symptoms to facilitate comparison between these groups. The EORTC QLQ-C30 and LC-13 were used for symptom and HRQOL assessments at baseline and at regular intervals up to 1 year (N=395).

RESULTS

NS patients had significantly more favourable baseline characteristics when compared to S patients with a median survival of 11.8 versus 6.0 months (P<0.0001), respectively. At baseline, S patients demonstrated HRQOL scores inferior to those of NS patients (P<0.01) for most scales. Until week 14, NS patients developed more symptoms while S patients experienced symptom relief in most scales. After week 14, no significant differences could be observed between the groups.

CONCLUSION

This study indicates that immediate TRT, given to patients with minimal/none chest symptoms, does not prevent development of disease-related symptoms and diminished HRQOL. A wait-and-see policy appears to be acceptable.

A prospective study of health-related quality of life, fatigue, anxiety and depression 3-5 years after stem cell transplantation

Health-related quality of life (HRQOL), fatigue and psychological distress were prospectively assessed in 248 cancer patients treated with allogeneic (SCT, N=61), or autologous (ASCT, N=69) stem cell transplantation or conventional chemotherapy (CT, N=118) of whom 128 completed the assessments after 3 years. The European Organization for Treatment and Research of Cancer Core Quality of Life Questionnaire and the Hospital Anxiety and Depression Scale were administered nine (SCT/ASCT groups) or seven times (CT group) during the first year. The Fatigue Questionnaire was added at the final assessment. The SCT group displayed greater changes from baseline scores than the ASCT group, with more symptoms in the first months post transplant. A gradual improvement was found in both groups during the following 4-6 months, before stabilizing at baseline levels. Only minor changes were observed after the first year. All groups reported more fatigue than the population values after 3 years (P<0.01). The ASCT group also reported less optimal HRQOL (P<0.01-0.0001). No differences were found in anxiety and depression. Despite a faster recovery during the first months after transplant, the ASCT patients reported poorer functioning and more fatigue compared to the SCT group after 3 years. This suggests a need for a closer follow-up of these patients with special emphasis on functional status and fatigue.

Quality-of-life measurement in clinical trials--the impact of causal variables

The theory of measurement scales, and in particular multi-item scales, has been extensively developed in educational testing, psychometric testing, personality testing, and consumer research. These scales are usually either based upon traditional psychometric models or modern theory using item response theory. However, clinical measuring instruments, including health-related quality-of-life questionnaires, frequently have different underlying principles and so the adoption of such approaches can be inappropriate. The fundamental statistical distinction between indicator and causal variables can be used to explain why psychometric methods fail. So-called clinimetric approaches may sometimes be more relevant, and clinimetric and psychometric ideas should be combined to yield a suitable measuring instrument. Recognition of the role of causal variables enables informed decisions to be made regarding scale development, validation, and scoring.

The CARES-SF used for prospective assessment of health-related quality of life after stem cell transplantation

OBJECTIVE

By employing the Cancer Rehabilitation and Evaluation System short form (CARES-SF) prospectively we wanted to focus on the rehabilitation needs after high-dose chemotherapy (HDC) and stem cell transplantation, in order to identify problems that should be addressed by health-care professionals during the course of disease and treatment.

METHODS

The CARES-SF was administered before and at 2, 6 and 12 months post-transplant to 130 cancer patients treated with HDC and allogeneic (SCT) or autologous stem cell transplantation (ASCT). Physical function scale scores were compared with the corresponding scale of the EORTC QLQ-C30.

RESULTS

The SCT group reported significantly better physical function than the ASCT group before transplant on both the CARES-SF (p<0.0001) and the EORTC QLQ-C30 (p<0.01). Almost identical mean CARES-SF scores across groups (SCT: 0.7-1.4, ASCT: 0.8-1.3) were found at the subsequent assessments, consistent with the QLQ-C30 data. Correlations between CARES-SF and QLQ-C30 Physical Function Scales ranged from 0.45 to 0.65. The SCT group had better psychosocial subscale scores (mean 0.4 and 0.5 versus ASCT: 0.7 and 0.8, p < 0.01) at the 6 and 12-month assessments, as well as better satisfaction on the marital subscale ( p=0.01) 6 months post-transplant. Few patients requested specific help: 19% at baseline with 'fear of the cancer progressing' and 9% with 'reduction in physical energy' after 6 and 12 months.

CONCLUSION

The CARES-SF detected differences across groups of patients as well as within-patient changes over time. The possibility for patients to express their need for professional assistance renders the CARES-SF appropriate after SCT/ASCT. The sexual, marital and medical interaction subscales in particular address specific issues of relevance for follow-up care, compared with more traditional questionnaires assessing health related quality of life (HRQOL).

Hypofractionated palliative radiotherapy (17 Gy per two fractions) in advanced non-small-cell lung carcinoma is comparable to standard fractionation for symptom control and survival: a national phase III trial

PURPOSE

To investigate whether the effect of hypofractionated thoracic radiotherapy (TRT) is comparable to more standard fractionated radiotherapy (RT) in advanced non-small-cell lung cancer (NSCLC).

PATIENTS AND METHODS

A total of 421 patients with locally advanced stage III or stage IV NSCLC tumors were included. Inclusion criteria were inoperable, disease too advanced for curative radiotherapy, and chest symptoms or central tumor threatening the airways. Patients were randomly assigned to three arms: A, 17 Gy per two fractions (n = 146); B, 42 Gy per 15 fractions (n = 145); and C, 50 Gy per 25 fractions (n = 130). Four hundred seven patients were eligible for the study; 395 patients (97%) participated in the health-related quality-of-life (HRQOL) study. The European Organization for Research and Treatment of Cancer (EORTC) Quality of Life Questionnaire (QLQ)-C30 and EORTC QLQ-lung cancer-specific module (LC13) were used to investigate airway symptom relief and changes in HRQOL. Assessments were performed before TRT and until week 54. Clinicians' assessments of symptom improvement were at 2, 6, and 14 weeks after completion of TRT. The patients were observed for a minimum of 3 years. Results Baseline prognostic data were equally distributed in the treatment groups. Patient compliance with respect to the HRQOL investigation was minimum 74%. HRQOL and symptom relief were equivalent in the treatment arms. No significant difference in survival among arms A, B, and C was found, with median survival 8.2, 7.0, and 6.8 months, respectively.

CONCLUSION

Our data indicate that protracted palliative TRT renders no improvement in symptom relief, HRQOL, or survival when compared with short-term hypofractionated treatment in advanced NSCLC.

Fetal, infant, and childhood growth and adult blood pressure: a longitudinal study from birth to 22 years of age

BACKGROUND

People who are small at birth tend to have higher blood pressure in later life. However, it is not clear whether it is fetal growth restriction or the accelerated postnatal growth that often follows it that leads to higher blood pressure.

METHODS AND RESULTS

We studied blood pressure in 346 British men and women aged 22 years whose size had been measured at birth and for the first 10 years of life. Their childhood growth was characterized using a conditional method that, free from the effect of regression to the mean, estimated catch-up growth. People who had been small at birth but who gained weight rapidly during early childhood (1 to 5 years) had the highest adult blood pressures. Systolic pressure increased by 1.3 mm Hg (95% CI, 0.3 to 2.3) for every standard deviation score decrease in birth weight and, independently, increased by 1.6 mm Hg (95% CI, 0.6 to 2.7) for every standard deviation score increase in early childhood weight gain. Adjustment for adult body mass index attenuated the effect of early childhood weight gain but not of birth weight. Relationships were smaller for diastolic pressure. Weight gain in the first year of life did not influence adult blood pressure.

CONCLUSIONS

Part of the risk of adult hypertension is set in fetal life. Accelerated weight gain in early childhood adds to this risk, which is partly mediated through the prediction of adult fatness. The primary prevention of hypertension may depend on strategies that promote fetal growth and reduce childhood obesity.

Cluster-randomized trials

Cluster-randomized trials represent an important experimental design, supplementing ordinary randomized clinical trials. They are particularly relevant when evaluating interventions at the level of clinic, hospital, district or region. They are necessary when it is not feasible to randomize individual patients, and desirable when there may be contamination between clusters. But they also carry serious design and analysis implications, and the use of clusters as the unit of randomization must be justified. Sample sizes will usually need to be greatly increased, an adequate number of clusters is essential, and the statistical analysis must allow for the cluster design. And one should rigorously guard against selection bias.

Lack of concealment may lead to selection bias in cluster randomized trials of palliative care

Comprehensive palliative care programs are often implemented on a community level, and to evaluate such interventions, randomization by cluster (community) may be the only feasible method. In trials randomizing individual subjects, the importance of proper concealment has been stressed. In cluster randomized trials, however, concealment of individual patient allocation is often impossible. The following risk of selection bias has been given little attention. In the present study, comparing palliative care to conventional care, community health care districts were defined as clusters and randomized. The patients' treatment assignment was determined by the allocation of the cluster in which they resided, and hence predictable by their address. A biased selection based on practical considerations related to patients' diagnoses and hospital departments was suspected. To explore this, cancer diagnoses were grouped according to local tradition for sharing of treatment responsibility among hospital departments. A significant difference between trial arms in distribution of these groups was revealed and strongly supported our suspicion. The finding carries an important message to future researchers: when using cluster randomization, any evidence of selection bias should be carefully checked and reported.

Analysis of messy longitudinal data from a randomized clinical trial. MRC Lung Cancer Working Party

The randomized clinical trial, LU19, conducted by the Medical Research Council Lung Cancer Working Party, was designed to compare ACE (doxorubicin, cyclophosphamide and etoposide) chemotherapy plus G-CSF (granulocyte colony-stimulating factor) at 2-week intervals versus ACE chemotherapy alone at standard 3-week intervals in patients with small-cell lung cancer. This trial investigated whether more intensive administration of ACE would improve overall survival and affect the quality of life of patients. The report on overall survival and other outcome measures will be published in the Journal of Clinical Oncology. In this paper we focus on methods of analysing aspects of data reflecting quality of life. Twelve symptoms of lung cancer and its treatment - cough, haemoptysis, pain, nausea, vomiting, hoarse voice, sore mouth, rash, lethargy, lack of appetite, alopecia, and dysphagia - were scheduled to be assessed on seven occasions for the ACE arm and on eight occasions for the ACE+G-CSF arm by clinicians during the first 18 weeks of the treatment period. However, in practice the number of assessment forms completed per patient ranged from 1 to 9, and assessment time-points were very different from those planned. These 'messy' longitudinal data are explored by both a summary measure approach, in which experience of a symptom is summarized by a single value, and an extensive model-based statistical approach, which explicitly takes into account correlation within repeated measures. These analyses provide a clear picture of symptom comparisons between the two treatments. The application of various methods offers not only an approach to assessing the robustness of the results but also a basis for investigating reasons for inconsistency of results across methods. We conclude that except lethargy, which is worse in the ACE+G-CSF arm, all symptoms are similar across the two arms during the treatment period.

A 12 country field study of the EORTC QLQ-C30 (version 3.0) and the head and neck cancer specific module (EORTC QLQ-H&N35) in head and neck patients. EORTC Quality of Life Group

This study tests the reliability and validity of the European Organization for Research and Treatment of Cancer (EORTC) head and neck cancer module (QLQ-H&N35) and version 3.0 of the EORTC Core Questionnaire (QLQ-C30) in 622 head and neck cancer patients from 12 countries. The patients completed the QLQ-C30, the QLQ-H&N35 and a debriefing questionnaire before antineoplastic treatment or at a follow-up. 232 patients receiving treatment completed a second questionnaire after treatment. Compliance was high and the questionnaire was well accepted by the patients. Multitrait scaling analysis confirmed the proposed scale structure of the QLQ-H&N35. The QLQ-H&N35 was responsive to differences between disease status, site and patients with different Karnofsky performance status, and to changes over time. The new physical functioning scale (with a four-point response format) of version 3.0 of the QLQ-C30 was shown to be more reliable than previous versions. Thus, the QLQ-H&N35, in conjunction with the QLQ-C30, appears to be reliable, valid and applicable to broad multicultural samples of head and neck cancer patients.

The CUP trial: a randomized study analyzing the efficacy of high dose therapy and purging in low-grade non-Hodgkin's lymphoma (NHL)

BACKGROUND

The CUP trial was initiated to analyze the value of high-dose therapy and stem-cell transplantation and purging in patients with relapsed chemosensitive follicular NHL.

PATIENTS AND METHODS

After three cycles of chemotherapy responsive patients were randomized to either three more cycles of the same chemotherapy (C), high-dose therapy followed by autologous unpurged (U) or purged (P) stem-cell transplantation. Purging was performed using a cocktail of monoclonals. Pretransplant conditioning consisted of cyclophosphamide (60 mg/kg x 2) and total body irradiation.

RESULTS

Of the 140 patients registered from 26 centers in Europe, 89 fulfilled the criteria for randomization (C: 24, U: 33 and P: 32). Reasons for failure to randomize were: no response (28), persistent marrow infiltration (4), patient refusal (7), other (7), no data (5). With the current follow up (median 26 months from randomization) 16 (66%) in C are known to have progressed or relapsed, in contrast to 13 (39%) of U and 12 (37%) of the P patients (P-value 0.002). Overall survival is premature with the current available data.

CONCLUSIONS

Patients in U and P arms had higher progression/relapse-free survival rate. There are some suggestions of some improvement in overall survival rate.