Baseline health-related quality-of-life data as prognostic factors in a phase III multicentre study of women with metastatic breast cancer

Issues and challenges with integrating patient-reported outcomes in clinical trials supported by the National Cancer Institute-sponsored clinical trials networks

PURPOSE

The objective of this report is to provide a historical overview of and the issues and challenges inherent in the incorporation of patient-reported outcomes (PROs) into multinational cancer clinical trials in the cancer cooperative groups.

METHODS

An online survey of 12 cancer cooperative groups from the United States, Canada, and Europe was conducted between June and August of 2006. Each of the cooperative groups designated one respondent, who was a member of one of the PRO committees within the cooperative group.

RESULTS

There was a 100% response rate, and all of the cancer clinical trial cooperative groups reported conducting PRO research. PRO research has been conducted in the cancer cooperative groups for an average of 15 years (range, 6 to 30 years), and all groups had multidisciplinary committees focused on the design of PRO end points and the choice of appropriate PRO measures for cancer clinical trials. The cooperative groups reported that 5% to 50% of cancer treatment trials and an estimated 50% to 75% of cancer control trials contained PRO primary and secondary end points. There was considerable heterogeneity among the cooperative groups with respect to the formal and informal policies and procedures or cooperative group culture towards PROs, investigator training/mentorship, and resource availability for the measurement and conduct of PRO research within the individual cooperatives.

CONCLUSION

The challenges faced by the cooperative groups to the incorporation of PROs into cancer clinical trials are varied. Some common opportunities for improvement include the adoption of standardized training/mentorship mechanisms for investigators for the conduct of PRO assessments and data collection and the development of minimal criteria for PRO measure acceptability. A positive cultural shift has occurred in most of the cooperative groups related to the incorporation of PROs in clinical trials; however, financial and other resource barriers remain and need to be addressed.

The prognostic effects of performance status and quality of life scores on progression-free survival and overall survival in advanced ovarian cancer

OBJECTIVE

Performance status (PS) is an important prognostic factor in advanced ovarian cancer. The purpose of this study was to evaluate the prognostic significance of PS and quality of life (QoL) assessment on progression-free survival (PFS) and overall survival (OS) in patients with advanced ovarian cancer.

METHODS

We studied Canadian patients participating in an intergroup study in ovarian cancer (NCIC-OV10), which randomized patients to receive either standard chemotherapy using cisplatin/cyclophosphamide or cisplatin/paclitaxel chemotherapy. QoL was assessed using the EORTC quality of life questionnaire (QLQ-C30+3). The effects of multiple variables including the relevant clinical variables, PS and QoL scores were analyzed by Cox stepwise regression at baseline and again 3 months after completion of chemotherapy.

RESULTS

At baseline and at 3 months after chemotherapy, there were 151 and 93 patients respectively who completed the QLQ-C30+3 questionnaires. Baseline PS, global QoL score and treatment were independent predictors for both PFS and OS. Baseline cognitive functioning score was also an additional independent predictor for OS. At 3 months after completion of chemotherapy global QoL score, PS and grade were significant independent predictors of OS; however, only physical functioning score, emotional functioning score and tumor grade predicted for PFS.

CONCLUSIONS

Performance status and global quality of life scores at baseline are prognostic factors in advanced ovarian cancer for both PFS and OS. Higher baseline cognitive functioning scores were also associated with improved survival. Global QoL scores at 3 following completion of chemotherapy proved to be of prognostic significance for OS but not PFS.

Pretreatment quality of life predicts for locoregional control in head and neck cancer patients: a radiation therapy oncology group analysis

PURPOSE

To analyze the prospectively collected health-related quality-of-life (HRQOL) data from patients enrolled in two Radiation Therapy Oncology Group randomized Phase III head and neck cancer trials (90-03 and 91-11) to assess their value as an independent prognostic factor for locoregional control (LRC) and/or overall survival (OS).

METHODS AND MATERIALS

HRQOL questionnaires, using a validated instrument, the Functional Assessment of Cancer Therapy-Head and Neck (FACT-H&N), version 2, were completed by patients before the start of treatment. OS and LRC were the outcome measures analyzed using a multivariate Cox proportional hazard model.

RESULTS

Baseline FACT-H&N data were available for 1,093 patients and missing for 417 patients. No significant difference in outcome was found between the patients with and without baseline FACT-H&N data (p = 0.58). The median follow-up time was 27.2 months for all patients and 49 months for surviving patients. Multivariate analyses were performed for both OS and LRC. Beyond tumor and nodal stage, Karnofsky performance status, primary site, cigarette use, use of concurrent chemotherapy, and altered fractionation schedules, the FACT-H&N score was independently predictive of LRC (but not OS), with p = 0.0038. The functional well-being component of the FACT-H&N predicted most significantly for LRC (p = 0.0004).

CONCLUSIONS

This study represents, to our knowledge, the largest analysis of HRQOL as a prognostic factor in locally advanced head and neck cancer patients. The results of this study have demonstrated the importance of baseline HRQOL as a significant and independent predictor of LRC in patients with locally advanced head and neck cancer.

Quality-of-life in elderly patients with cancer: a short review

BACKGROUND

Prolongation of survival and maintenance or improvement of health-related quality-of-life (HRQoL) are the two important goals within the treatment of individual patients. Due to the severity of symptoms and the toxicity of treatment, HRQoL has become a major area of concern when treating cancer patients in general and elderly patients in particular.

PATIENTS AND METHODS

We present a literature review of HRQoL aspects in elderly patients with cancer and especially address the topic whether impairments in the different tools of a comprehensive geriatric assessment (CGA) are associated with decreased HRQoL in elderly cancer patients.

RESULTS

Elderly cancer patients tend to weight their HRQoL as more important than gain in survival, when compared to younger patients. An age-dependent decrease in different scales of HRQoL is reported in patients and normative samples. HRQoL is also a predictor of survival. The variation of HRQoL can be used in trials comparing different treatment options. In individual patients, regular measurement of HRQoL aims to improve patients-centred care. Age related impairments of different areas of CGA are associated with decreased HRQoL in elderly cancer patients.

CONCLUSIONS

HRQoL is an important outcome with elderly cancer patients and should be assessed regularly and thoroughly.

The prognostic value of health-related quality-of-life data in predicting survival in glioblastoma cancer patients: results from an international randomised phase III EORTC Brain Tumour and Radiation Oncology Groups, and NCIC Clinical Trials Group study

This is one of the few studies that have explored the value of baseline symptoms and health-related quality of life (HRQOL) in predicting survival in brain cancer patients. Baseline HRQOL scores (from the EORTC QLQ-C30 and the Brain Cancer Module (BN 20)) were examined in 490 newly diagnosed glioblastoma cancer patients for the relationship with overall survival by using Cox proportional hazards regression models. Refined techniques as the bootstrap re-sampling procedure and the computation of C-indexes and R²-coefficients were used to try and validate the model. Classical analysis controlled for major clinical prognostic factors selected cognitive functioning (P=0.0001), global health status (P=0.0055) and social functioning (P<0.0001) as statistically significant prognostic factors of survival. However, several issues question the validity of these findings. C-indexes and R²-coefficients, which are measures of the predictive ability of the models, did not exhibit major improvements when adding selected or all HRQOL scores to clinical factors. While classical techniques lead to positive results, more refined analyses suggest that baseline HRQOL scores add relatively little to clinical factors to predict survival. These results may have implications for future use of HRQOL as a prognostic factor in cancer patients.

Coping with recurrent breast cancer: predictors of distressing symptoms and health-related quality of life

Little is known about how postmenopausal women with recurrent breast cancer cope with distressing symptoms and which factors predict health-related quality of life (HRQOL). In the present study, 56 consecutively enrolled patients completed questionnaires measuring symptom occurrence, coping capacity, coping efforts, and HRQOL at the time of recurrence. Results from this study illustrate that women with recurrent breast cancer suffer from multiple, concurrent, and interrelated symptoms of illness, anxiety, depression, and fatigue. Highly prevalent symptoms are lack of energy, difficulty sleeping, pain, worrying, problems with sexual interest, feeling sad, and dry mouth. The most frequently occurring symptom is problem with sexual interest, and the most severe symptom is worrying. The most distressing symptom experienced is pain. The majority of the women report 10-23 symptoms. Women who experience multiple symptoms also report higher levels of symptom distress. The experience of distressing symptoms is predicted by coping capacity, and the coping efforts experienced predict HRQOL. Patients with lower coping capacity report higher prevalence of symptoms, experience higher levels of distress, and experience worse perceived health, which in turn may decrease their HRQOL. To help women manage recurrent breast cancer, it is important to use multidimensional measurement to identify, evaluate, and treat distressing symptoms, and not assess single symptoms only. Care must be based upon the awareness of critical factors that exacerbate vulnerability to distress, as well as the ability to adapt to a recurrent breast cancer disease.

Quality of life as a predictor of cancer survival among Chinese liver and lung cancer patients

The utility of quality of life (QoL) scores in predicting cancer survival remains inconclusive because of methodological and/or statistical heterogeneity. We examined whether QoL scores predicted survival among Chinese liver (n=176) and lung cancer (n=358) patients. Cox proportional hazards models examined if QoL and psychosocial variables predicted survival after fully adjusting for sociodemographic and clinical factors. The results showed that global QoL scores did not predict survival in either patient group. Less advanced cancer stage (HR=2.574, p<0.05) was associated with longer survival in liver cancer. Longer survival in lung cancer was predicted by younger age (HR=1.016, p<0.05), less advanced cancer stage (HR=1.978, p<0.001), having had treatment before baseline (HR=0.671, p<0.05), better physical well-being (HR=0.941, p<0.001) and better appetite (HR=0.888, p<0.001). Global QoL (FACT-G(Ch)) scores do not predict survival in Chinese liver and lung cancer patients. QoL physical well-being subscale predicted lung cancer survival.

Impact of chemotherapy beyond the first line in patients with metastatic breast cancer

PURPOSE

The goal of this study was to determine which benefit could be brought by the succession of chemotherapy lines in patients treated for metastatic breast cancer and to identify patients who benefit from these treatments.

PATIENTS AND METHODS

Nine hundred and thirty four patients with metastatic breast cancer diagnosed between 1992 and 2002 were studied. A total of 772, 505, 283, 127 and 55 patients received a first, second, third, fourth and fifth line of chemotherapy, respectively. The evaluation of benefit in each chemotherapy lines was based on time of disease control (TDC). TDC is defined as time between the date of the beginning of treatment and the date of progression of disease or death. In case of early progressive disease or progression diagnosed at the first evaluation, TDC value is zero. Cox proportional hazards model was used to identify factors that could influence the TDC in each line of chemotherapy.

RESULTS

Medians duration of TDC are 9.3 [0-120], 5.9 [0-83.6], 4.63 [0-37.2], 4.1 [0-36.7] and 0.23 months [0-15] in first, second, third, fourth and fifth lines, respectively. More interestingly, TDC was longer than 6 months in 50.5% of patients treated with second line, 40% with third line, 35% with fourth line and 23.5% with fifth line. These results justify the therapeutic succession of chemotherapy lines in metastatic breast cancer. Only one factor influences the duration of TDC for each line of treatment: duration of TDC observed in the previous line.

Psychological distress and fatigue predicted recurrence and survival in primary breast cancer patients

This study examined whether psychological distress in newly diagnosed breast cancer patients was associated with their survival. We analyzed data from 1,588 breast cancer patients who filled in the EORTC QLQ-C30 questionnaire and the Hospital Anxiety and Depression Scale (HADS) 2 months after their primary operation. The median follow-up time was 12.9 years. Psychological distress (EORTC QLQ-C30 emotional function; HADS anxiety; HADS depression) and EORTC fatigue, physical function, and overall ratings were used to predict recurrence-free and overall survival, controlling for the known clinical and histopathological prognostic factors (biological model) using Cox multivariate regression analysis. Low levels of psychological distress (good EORTC emotional function) and low fatigue independently predicted longer recurrence-free and overall survival, controlling for biological factors. Lack of anxiety (HADS) also predicted longer recurrence-free survival. When added in combination to the biological model, fatigue remained a significant predictor of recurrence-free survival (P = 0.0004; risk ratio 1.32 (1.13-1.54)) and emotional function remained a significant predictor of overall survival (P = 0.0074; risk ratio 0.81 (0.70-0.95)). Low psychological distress and a low level of fatigue may cause a greater cancer resistance or may reflect underlying mental and physical robustness.

The anti-cancer drug-induced pica in rats is related to their clinical emetogenic potential

Cancer chemotherapy is frequently accompanied by severe emesis. The anti-cancer drugs are classified according to their clinical emetogenic potential. We have already found that kaolin ingestion behavior "pica" is analogous to emesis in rats. The aim of this study was to examine the effects of the clinical emetogenic potential of anti-cancer drugs on the induction of the pica in rats. Rats were housed in individual cages with free access to food and kaolin pellets and the daily food and kaolin intakes were measured for 3 days after the intraperitoneal administration of anti-cancer drugs (cisplatin, cyclophosphamide, actinomycin D, 5-fluorouracil and vincristine). The drugs with high potential for inducing emesis, such as cisplatin and cyclophosphamide, induced pica in all animals on the day of administration and the behavior lasted during the observation period. The drugs with moderate emetogenic potential, i.e. actinomycin D and 5-fluorouracil, also induced pica on the first and second day after the drug administration but the kaolin intake was less than that of the drugs with high potential. Vincristine, a drug with low emetogenic potential, slightly increased the kaolin intake in rats on the only first day of the administration. Cyclophosphamide, actinomycin D and vincristine induced anorexia and decreased their body weight during the observation period. These results suggested that the both amounts of kaolin intake and duration of behavior in the anti-cancer drug-induced pica are related to the clinical emetogenic potential of the drugs and the incidence of the anorexia is not related to their emetogenic potential.

The prognostic association of health-related quality of life scores with survival in breast cancer

GOALS OF WORK

There is extensive data showing that health-related quality of life (HRQOL) tools measuring the activities of daily life provide prognostic information in cancer. However, similar information on HRQOL tools measuring patient satisfaction with their life is sparse. The Ferrans and Powers quality of life index (QLI) is one such instrument. This study evaluated the association between HRQOL, as measured by the QLI, and survival in breast cancer.

MATERIALS AND METHODS

We examined a consecutive case series of 251 histologically confirmed breast cancer patients treated at Cancer Treatment Centers of America(R) between April 2001 and November 2004. QLI measures overall HRQOL and the HRQOL in four major subscales: health and physical functioning, social and economic, psychological and spiritual, and family. All scores range from 0 to 30 with higher scores indicating a better HRQOL. Study patients were dichotomized into two groups based on the median scores for all QLI subscales. Kaplan-Meier method was used to calculate survival. Log-rank test was used to study the equality of survival distributions. Multivariate Cox regression analyses were then performed to evaluate the joint prognostic significance of HRQOL and clinical factors.

RESULTS

Patient satisfaction with health and physical subscale was significantly associated with survival (p = 0.0006), with the median survival for low and high scores being 17.8 and 35.3 months, respectively. Similarly, patient satisfaction with overall HRQOL was significantly associated with survival (p = 0.0006), with the median survival for low and high scores being 17.8 and 34.6 months, respectively. Patient satisfaction with health and physical subscale and overall HRQOL were found to be predictive of survival independent of the effects of stage at presentation.

CONCLUSIONS

This study suggests that baseline patient satisfaction with health and physical functioning and overall HRQOL, as measured by QLI, provides useful prognostic information in patients with breast cancer independent of stage at presentation.

Prognostic significance of symptoms of hospitalised advanced cancer patients

PURPOSE

To assess the prognostic value of symptoms in hospitalised advanced cancer patients.

PATIENTS AND METHODS

A prospective analysis was performed of 181 hospitalised patients referred to a Palliative Care Team. Comprehensive symptom questionnaire, functional status, estimated life expectancy and survival were assessed. Using a Cox regression model, a predictive survival model was built.

RESULTS

Median survival: 53 d. Median number of symptoms: 4; 20 symptoms occurred in 10%. Multivariate analysis showed nausea, dysphagia, dyspnoea, confusion and absence of depressed mood as independent prognostic factors for survival (p<0.05) with relative risks of dying of 1.96, 1.81, 1.79, 2.35 and 1.79, respectively. Patients with 2, 3 or 4 of these factors at the same time had a relative risk of dying of 2.7, 2.1 and 9.0, respectively.

CONCLUSION

A cluster of factors comprising nausea, dysphagia, dyspnoea, confusion and absence of depressed mood may be used to accurately predict survival in hospitalised advanced cancer patients.

Can patient satisfaction with quality of life predict survival in advanced colorectal cancer?

GOALS OF WORK

The goal of this study was to evaluate the association between patient satisfaction with quality of life (QoL) and survival in colorectal cancer patients undergoing care in a community hospital comprehensive cancer center.

MATERIALS AND METHODS

A consecutive case series of 177 cases of histologically confirmed colorectal cancer treated at Cancer Treatment Centers of America at Midwestern Regional Medical Center between April 2001 and November 2004. The quality of life index (QLI) was utilized to assess patient satisfaction with QoL. QLI measures global QoL and the QoL in four major subscales: health and physical functioning, social and economic, psychological/spiritual, and family. All scores range from 0 to 30 with higher scores indicating a better QoL. Kaplan-Meier method was used to calculate survival. Log-rank test was used to study the equality of survival distributions. Multivariate Cox regression analyses were then performed to evaluate the joint prognostic significance of those QoL and clinical factors that were shown to be prognostic in univariate analyses.

RESULTS

Of the 177 patients, 46 were newly diagnosed and 131 have treatment history. The median age was 53 years (range 25-85 years). Eight patients had stage I disease, 16 had stage II, 51 had stage III, and 77 had stage IV. Health and physical subscale was significantly associated with survival (p=0.0003), with the median survival for low scores being 8.3 and 20.6 months for high scores. Health and physical subscale was found to be predictive of survival independent of the effects of tumor stage at diagnosis and treatment history.

CONCLUSIONS

We found that baseline patient satisfaction with QoL, as measured by the QLI, provides useful prognostic information in patients with colorectal cancer independent of tumor stage at diagnosis and treatment history. While these findings require further investigation in large patient cohorts, they may have important implications for patient stratification in clinical trials and aid in clinical decision-making.

Quality of life is predictive of survival in patients with unresectable hepatocellular carcinoma

BACKGROUND

Patients with unresectable hepatocellular carcinoma (HCC) have a dismal prognosis. The objective of this study was to evaluate whether patient-reported baseline quality of life (QoL) measured by the EORTC QLQ-C30 instrument is predictive of survival for these patients.

MATERIALS AND METHODS

Two hundred and thirty-three patients with unresectable HCC (mainly hepatitis B-associated) who were recruited into two separate randomized phase III clinical studies, based on palliative chemotherapy and palliative hormonal therapy, respectively, gave consent and received pretreatment QoL assessment. EORTC QLQ-C30 scores and clinical variables at the time of study entry were analyzed to identify factors that influenced survival by applying multivariate analysis. Independent prognostic factors for survival were studied by Cox regression analysis.

RESULTS

Median survival of the 233 patients was 5.5 months (95% CI 4.2-6.5 months). Significant independent predictors of shorter survival were advanced Okuda staging (P = 0.0030; HR = 2.058), high baseline total bilirubin (P = 0.0008; HR = 1.013) and worse QoL score in the appetite score domain (P = 0.0028; HR for 10 point increase = 1.070). Patients who were entered into the chemotherapy trial (P = 0.0002; HR = 0.503), those who scored better in the physical functioning domain (P = 0.0034; HR for 10 point decrease = 0.911) and the role functioning domain (P = 0.0383; HR for 10 point decrease = 0.944) of the QoL questionnaire, were associated with longer survival.

CONCLUSIONS

In the studied HCC population, patient-reported baseline QoL provides additional prognostic information that supplements traditional clinical factors, and is a new prognostic marker for survival for patients with unresectable HCC.

Does a patient’s self-reported health-related quality of life predict survival beyond key biomedical data in advanced colorectal cancer?

The purpose of this study was to determine whether baseline patients' self reported health-related quality of life (HRQOL) parameters could predict survival beyond key biomedical prognostic factors in patients with metastatic colorectal cancer. The analysis was conducted on 299 patients. HRQOL baseline scores were assessed using the European Organisation for Research and Treatment of Cancer, Quality of Life Questionnaire-Core30 (EORTC QLQ-C30). The Cox proportional hazards regression model was used for both univariate and multivariate analyses of survival. In addition, a bootstrap resampling technique was used to assess the stability of the outcomes. The final multivariate Cox regression model retained four variables as independent prognostic factors for survival: white blood cell (WBC) count with a hazard ratio (HR) of 1.961 (95% CI, 1.439-2.672; P<0.001), alkaline phosphatase with HR=1.509 (95% CI, 1.126-2.022; P=0.005), number of sites involved with HR=1.108 (95% CI, 1.024-1.198; P=0.01) and the patient's score on the social functioning scale with HR=0.991 (95% CI, 0.987-0.996; P<0.001) which translates into a 9% decrease in the patient's hazard of death for any 10 point increase. The independent prognostic importance of social functioning and the stability of the final Cox regression model were also confirmed by the additional bootstrap model averaging analysis, based on 1000 bootstrap-generated samples. The results suggest that social functioning, acts as a prognostic measure of survival beyond a number of previously known biomedical parameters.

Health related quality of life outcomes in cancer clinical trials

Over the last decade, health related quality of life (HRQOL) investigations have become an increasingly important part of many cancer clinical trial research programs. This paper presents a review of all HRQOL studies published by the European Organisation for Research and Treatment of Cancer (EORTC), one of the largest clinical trials organisations in Europe. The findings highlight 24 clinical trials that have been published to date, enrolling over 9000 patients. HRQOL is fully integrated into EORTC phase III trials. In many trials, HRQOL provides a valuable source of additional information useful to both clinician and patient when making treatment decisions. Furthermore, several trials have found that the combined use of clinical information along with HRQOL data has led to the development of new standards of care in several different cancer sites. With more than 40 ongoing HRQOL studies in the EORTC, we expect HRQOL to play an even greater role over the coming decade in helping establish the optimal treatment and care approach for cancer patients.

Health-Related Quality of Life Parameters As Prognostic Factors in a Nonmetastatic Breast Cancer Population: An International Multicenter Study

Purpose

The purpose of this research was to evaluate whether baseline health-related quality of life (HRQOL) parameters are prognostic factors for survival in locally advanced breast cancer patients. Although the literature highlights the important role of HRQOL parameters in predicting survival in advanced metastatic disease, little evidence exists for earlier stages.

Patients and Methods

The overall sample consisted of 448 patients randomly assigned to receive cyclophosphamide, epirubicin, and fluorouracil versus epirubicin, cyclophosphamide, and granulocyte colony-stimulating factor. Patients were enrolled in 12 countries. HRQOL baseline scores were assessed using the European Organization for Research and Treatment of Cancer Quality of Life Questionnaire C30. The Cox proportional hazards regression model was used for both univariate and multivariate analyses of survival. In addition, a bootstrap resampling technique was used to assess the stability of the outcomes. Bootstrap results were then applied for model averaging purposes as a means to account for the observed model selection uncertainty.

Results

The final multivariate model retained inflammatory breast cancer (T4d) as the only factor predicting overall survival (OS) with a hazard ratio of 1.375 (95% CI, 1.027 to 1.840; P = .03). The presence of inflammatory breast cancer lowers the median survival time from 6.6 to 4.2 years (36% reduction). None of the preselected HRQOL variables were prognostic for OS or disease-free survival, in either the univariate or multivariate analysis.

Conclusion

Our findings suggest that baseline HRQOL parameters have no prognostic value in a nonmetastatic breast cancer population.