| Japanese Journal of Clinical Oncology | Pages |
Editorial
RESEARCH IS LONG, LIFE IS SHORT
Medical oncologists, like a Modern Hippocrates, are struggling to study patients’ Quality-of-Life (QOL) in cancer chemotherapy. We frequently encounter scepticism of physicians, such as ‘one man’s meat is another man’s poison’, in conducting a QOL study in a large-scale multi-center trial setting. Some oncologists criticize the acquisition of QOL data as time consuming, saying ‘Good was good and bad was bad, so what? There will be nothing more than a garbage-pile of the data’. In spite of various oppositions, QOL research in cancer medicine has become one of the major scientific research fields with efforts to resolve many obstacles such as compliance of data acquisition and establishment of statistical methods for analysis.
In this issue, Han and co-workers (1) challenged to this difficult matter of the QOL assessment of chemotherapy for advanced non-small cell lung cancer in a palliative single-arm phase II setting. It is sometimes puzzling for us to evaluate QOL data of the cohort in a single-arm phase II study. Small sample size, large variability of the patients’ background and patient selection will cause unavoidable bias and there may be pitfalls on judgment of the outcome of the treatment. Nevertheless, evaluation of patient’s perception for toxicity is an important research area. In fact, discrepancies between physicians and patients in rating toxicity degree and treatment benefits were found in many previous studies. According to the article ‘outcomes of cancer treatment for technology assessment and cancer treatment guideline by the Committee of American Society of Clinical Oncology’, QOL is an important endpoint because it reflects how patients feel (2). I have been concerned, in Han’s study, by the fact that the instrument was completed by attending physicians asking patients each item of questionnaires. This might have caused bias, because patients sometimes hesitate about being truthful to physicians and pretend to be better or worse than they actually are. The patient self-rating method, without intervention by anyway, will lessen this bias. Likewise, the differences in the backgrounds which were disclosed by the physicians and the perception of patients for their disease state might have also influenced the QOL change after treatment.
There are some important points in the set-up of a QOL study, such as instruments, timing and analytical methods (3). The questionnaires should be reasonably validated in comparison with other instruments and be short enough to be easily completed by patients. LASA is one of the instruments frequently used. According to the literature review by Montazeri (4), over 50 instruments were used to measure QOL in lung cancer studies. There are three lung cancer-specific measures with good reliability and validity: EORTC-QLQ, LCSS, FACT-L. Palliation of symptoms, psychosocial interventions, and understanding patients’ feelings and concerns contribute to improving QOL in lung cancer patients. A QOL study for evaluating treatment modality should be well-prepared before the start of the protocol. If not, it would be like putting a cart before a horse. To improve compliance of QOL data acquisition the study group and data management staff should be well-organized and trained.
What is the real outcome of the treatment? Outcomes are defined as the products, both good or bad, of cancer treatment according to the ASCO guideline mentioned above (2). Patient outcome is the effect of treatment on the patient, i.e. survival, toxicity and QOL. Regarding outcomes in Han’s study, the QOL at 6 weeks after MIP was shown to be better in this patients’ cohort. However, the compliance of the QOL measurement was only 27% of the patients at 12 weeks. It is well known that the impact of chemotherapy for advanced non-small cell lung cancer is actually very small. We have to be careful in judging from Han’s study that MIP regimen is good for palliation of symptoms and keeping patients’ QOL. The true benefit of this treatment may be small in terms of the trade-off relationship between benefits and toxicity. The results from this study on MIP regimen should be confirmed with a large-scale multi-center phase III study.
There were papers showing a correlation between pretreatment QOL values and patient survival in breast cancer, melanoma and lung cancer. In Han’s study the change in QOL scores was found to be an independent prognostic factor for survival. There were some argument about the inclusion of time-dependent factors as a variable in the analysis of prognostic factors. They applied some time-dependent factors, such as, steep decrease of serum tumor marker in an unit period after treatment as a predictor for good prognosis. In an ECOG 5592 trial, the QOL change after 6 weeks of treatment showed a significant correlation with patient survival (5). More studies will be needed to confirm the change of QOL to be a useful prognostic factor in lung cancer patients.
What new information has emerged about QOL research on cancer medicine in the past several years? Osoba (6) reviewed what we have learned from QOL studies in medical oncology and summarized as follows: (1) health-related QOL (HQL) is a multidimensional construct with multidimensional instruments; (2) the possibility to improve compliance in the collection of self-report HQOL data is real; (3) aggressive therapy may result in improved HQL; (4) symptoms are associated with quantifiable disruption in HQL; and 5) pretreatment HQL may be predictive of survival. Issues to be solved in future studies are as follows: the most appropriate instruments to achieve the purpose of measuring HQL; difference between statistically significant and clinically meaningful, or subjectively significant methods and validation of scoring weights or values; optimal timing of measurement; distinction of the effect on HQL from those of disease; and the ethical dilemma with possible abuse of QL information with various analytical methods.
Some treatment benefits have been shown in the meta-analysis of the randomized trials in patients with advanced non-small cell lung cancer compared to the best supportive care without any chemotherapy. Although the benefit may be small, patients with good risk should be proposed chemotherapy. Symptoms relief is one of the major objectives, and QOL measurement is an important method in evaluating the outcome of a treatment. However, in this single-arm phase II study it is very difficult to judge the real benefit of the regimen, because there were no comparisons with ‘the standard’, and various kinds of bias exist such as validation of the instrument, patient selection, small sample size, the timing of data collection, data attrition etc. One of the important purposes of the QOL study is to clarify which treatment modality can achieve better outcome on survival and toxicity, and large-scale randomized comparative trials are usually the best research field of the QOL study.
Recently, additional key papers focused on QOL research in cancer medicine have been published. Hjermstad et al. (7) reported a study on making standard the use of EORTC-QLQ C30 in the non-cancerous Norwegian population which was performed cross-sectionally. With this ‘standard’, physicians may be able to evaluate the outcome of intensive cancer treatment for each patient. Detmar et al. (8) performed a trial of QOL measurement in daily clinical oncology practice. Application of the QOL data to individual cancer patients has been an annoying problem for researchers, and in this study they discussed topics from the QOL data which emerged in an outpatient clinic. The physicians noticed some new topics that patients were worrying about and that they could discuss them more frequently with the patients. This showed a new insight to the future study on individualization of the QOL data probably beneficial for facilitating communication between doctors and patients in oncology practice. If it works, it will be very helpful for cancer patients who have anxiety about the daily life and still feel a lack of communication with the doctor. Cooperation of oncologists and QOL research staff including statistician is essential. In practice every cancer patient should discuss with physicians the trade-off relationship of the benefit and toxicity of the treatment. ‘While there is well-designed research, there is hope’ may be a real world in QOL research in cancer medicine.
References
|
Kenji Eguchi, M.D. Department of Internal Medicine and Medical Oncology National Shikoku Cancer Center |
This page is run by Oxford University Press, Great Clarendon Street, Oxford OX2 6DP, as part of the OUP Journals
Comments and feedback: www-admin{at}oup.co.uk
Last modification: 24 Nov 1998
Copyright©Japanese Journal of Clinical Oncology, 1998.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||