Improved QOL Considered To Be a Valid Clinical Endpoint in Itself

TAMPA, Florida—"Increasingly we have acknowledged that there is more to cancer treatment than the eradication of disease," stated Paul Jacobsen, PhD. "The treatment of anemia and its effects on quality of life have demonstrated that a treatment can be considered effective and clinically useful if it results in significant quality-of-life benefits, even in the absence of significant survival benefits."

The challenge, as outlined by Dr. Jacobsen, is how to accurately measure quality-of-life issues and incorporate them into clinical practice. Dr. Jacobsen is program leader, psychosocial and palliative care program, at the H. Lee Moffitt Cancer Center, and professor, department of psychology at the University of South Florida in Tampa.

It is commonly agreed that quality of life is "a multidimensional construct that is best measured using instruments that assess multiple domains of functioning and well-being," Dr. Jacobsen said. "By definition, it is a subjective phenomenon and the patient is the best judge of his or her own quality of life."

Three quality-of-life measures are commonly used in oncology:

• Functional Assessment of Cancer Therapy (FACT);

• European Organization for Research and Treatment of Cancer Quality-Of-Life Questionnaire (EORTC QLQC-30); and

• Linear analogue self-assessment scales (LASA).

What and When to Measure

Selection of quality-of-life measures for a study depends on the specific research questions and the populations being studied. While frequent assessments of quality of life would provide more complete data, constraints on frequency include not wanting to burden patients and the financial costs involved in collecting and compiling data.

Beneficial Uses of Data

Quality-of-life data have beneficial uses at the group level, Dr. Jacobsen observed. "First, it has allowed us to describe the effects of disease or treatment on functioning well-being. We now have a body of empirical data that can inform us as treatment providers and also patients about what to expect as they begin therapy."

The data also "allows us to identify the supportive and palliative care needs of patients," Dr. Jacobsen continued. "Some research has shown that quality-of-life measures seem to have prognostic significance or predictive value, certainly in predicting patients’ reactions to treatment in terms of toxicity. In the case of breast cancer and lung cancer, there is some evidence which suggests that quality of life is an independent prognostic indicator, over and above other factors known to be correlated with survival," he said.

"In clinical trials, we’ve been able to determine whether a treatment improves quality of life beyond pretreatment values. I think in the anemia story, the results with erythropoietin exemplify this. And in the instance where we have two treatments of similar survival value, we can look to quality-of-life data to decide which treatment might be preferable." Already "these data have been helpful in deciding among treatments for early-stage breast cancer and early-stage prostate cancer."

Helping Individual Patients

For individual patients, asking about quality-of-life concerns "can facilitate patient-provider communication," and "often leads to identifying frequently overlooked problems like fatigue, which we call the silent symptom," Dr. Jacobsen said. Questionnaires should be easy to read and take no longer than 5 minutes to complete, he recommended.

Models for Clinical Practice

Dr. Jacobson presented two models for using quality-of-life data in clinical practice. The first is based on a study in the Netherlands involving 6 physicians and 18 patients and using the EORTC QLQC-30 questionnaire at the baseline and two follow-ups (see Figure 1). The questionnaire took 5½ minutes to complete and 3 minutes to score and print so the results were quickly available to patients and physicians.

Physicians were more positive about using the questionnaire, with 72% reporting it useful, compared to 50% of patients. The patients expressed disappointment that physicians didn’t get into all the quality-of-life issues, suggesting that patients would have preferred more time than the average 8 minute visit to talk to the physicians.

For the second model, Dr. Jacobson chose the National Comprehensive Cancer Network (NCCN) universal screening practice guidelines for the management of psychological distress, pain, and fatigue.

Several symptom screening measures are available. Dr. Jacobsen said he personally favors the M.D. Anderson Symptom Inventory for screening because "it strikes a nice balance between being too long and too short.’’ (see Figure 2).

Unresolved Issues

Among the unresolved issues is how to use patient scores to define a problem and assist in clinical decision-making. Another issue involves distinguishing between screening and assessing. "It seems to be more economical and make more sense to use very brief measures to screen patients for the presence of symptoms and then use longer measures that focus perhaps on a single symptom to learn more about the nature of the problem," he said.

"I think the true promise of using quality-of-life data in clinical practice is when the results are tied with guidelines and clinical pathways into clinical decision-making and into applications of therapies," Dr. Jacobsen concluded.