Converting Quality of Life Data to 'Q' Scores Allows Comparisons

June 1, 1996

FORT LAUDERDALE, Fla--Although many quality of life measures for cancer have been validated, eg, the FLIC (Functional Living Index-Cancer) and the FACT (Functional Assessment of Cancer Therapy), use of a single instrument facilitates analysis of data obtained at different sites, David F. Cella, PhD, said at the first annual conference of the National Comprehensive Cancer Network (NCCN).

FORT LAUDERDALE, Fla--Although many quality of life measures forcancer have been validated, eg, the FLIC (Functional Living Index-Cancer)and the FACT (Functional Assessment of Cancer Therapy), use ofa single instrument facilitates analysis of data obtained at differentsites, David F. Cella, PhD, said at the first annual conferenceof the National Comprehensive Cancer Network (NCCN).

However, it is now possible to convert the total or aggregatescore from a specific assessment tool to a "Q" scorethat provides "a reasonable estimate of a person's qualityof life," he said, adding that Q score can also potentiallybe used as a modifier of survival time.

Dr. Cella and his colleagues at Rush Medical College, Chicago,where is he associate professor of psychology and internal medicine,have shown that the total scores obtained with FLIC or FACT (developedby Dr. Cella) are equivalent.

"This means that you can take the total score from eitherinstrument and convert it to a standardized Q score," hesaid, "and then use that standardized score as the basisfor discussion and comparison across trials, across physicians,or between physician and patient." The Rush researchers havedevised a table that converts a raw FACT score to a FLIC scoreand a Q score.

About a dozen subscales are available for use with the FACT-G,a 34-item general quality of life measure, targeted to specifictreatments, disease symptoms, issues (such as treatment satisfaction),and even protocols, Dr. Cella said.

The ovarian cancer subscale, for example, asks about problemsspecific to ovarian cancer such as abdominal swelling, cramps,mobility, weight loss, and bowel control. An ovarian cancer therapymight not be deemed effective measured by the FACT-G, he said,but could be shown to have a pronounced impact on symptom relief,measured by the subscale.

"The better you're able to chart quality of life benefits,the more ammunition you're going to have to justify a treatment,"Dr. Cella said, especially in the era of managed care. He saidthat physicians should not assume that the impact of treatmentson quality of life will be negative if the treatment has no measurableeffect on response rate and survival.

For example, cisplatinum-based chemotherapy produces partial responsesin 20% to 40% of patients with stage IV non-small-cell lung cancer,with a marginal survival benefit of about 10 to 15 weeks. Sucha small benefit makes it very difficult to justify giving thistherapy if survival is the only demonstrable endpoint, especiallyconsidering toxicity.

However, in the three phase II studies that have compared responsewith symptom improvement, in each case, symptom improvement wasfar better than response rate. "So there may be some subclinical,nonmeasurable response to treatment," Dr. Cella said.

Dr. Cella also cited two studies in which the increased toxicitythat accompanies intensive regimens did not negate quality oflife gains seen with treatment. At Princess Margaret Hospital,researchers found not only higher remission rates and increasedtoxicity but also improved quality of life in patients receivinga full dose of CMF, compared with low-dose CMF in advanced breastcancer.

"The dose-response relationship has to be evaluated withrespect to quality of life and toxicity together," Dr. Cellasaid. "It's neither safe nor fair to assume that you're measuringquality of life with the proxy of toxicity ratings, not only becausetoxicity and quality of life don't always go in the same direction,but also because toxicity is usually rated by providers and qualityof life by patients."

The Australian-New Zealand breast cancer study group found betterdisease response in patients with advanced breast cancer withcontinuous vs intermittent therapy, he said, ie, longer time toprogression, a trend toward improved survival, and improved qualityof life.

"They went into this study expecting the quality of lifeto be better with intermittent chemotherapy," Dr. Cella said,"the logic being that if a woman is asymp-tomatic, why giveher chemotherapy and add that burden of toxicity? Why not waituntil a problem occurs?"

But the hypothesis proved wrong, and this trial has become oneof the most frequently cited quality of life studies because ofits possible impact on practice. "Physicians who have paidattention to this study might be more likely to justify continuouschemotherapy even in asymptomatic advanced breast cancer,"he said.

Dr. Cella noted that patients have different ideas about treatment-relateddetriments in quality of life, with some having more tolerancefor side effects if even small survival gains are possible.

"Basically, Americans are very aggressive," he said.He described his study in which about 300 cancer patients wereasked how much treatment toxicity they would accept for an increasedchance of cure (early stage disease) or increased survival (advancedstage disease).

For early stage disease, patients were asked if they would accepttherapy with mild or severe toxicity if it would improve theirbaseline 50% chance for cure to 90%, meaning a 40% increment incure rate. "If they answered yes, we lowered the chance ofcure to 75%, meaning an improvement of 25%," Dr. Cella said.

He noted that the study was originally designed to stop at a 55%chance of cure "but so many people were accepting 55% thatwe had to go down to 51%." Some 80% of patients said theywould refuse therapy with mild toxicity for a 1% improvement inlikelihood of benefit, and 90% would refuse therapy with severetoxicity for this small benefit.

However, Dr. Cella pointed out another way to look at the finding:10% of people are still saying they want therapy for a 1% improvementin likelihood of cure, even if the treatment is severely toxic.The figure is 20% for mild toxicity.

In the advanced disease scenarios, patients were asked if theywould accept treatment with mild or severe toxicity if their 3-monthlife expectancy could be extended by 5 years. If they said yes,this was lowered to 1 year, then 6 months, then 1 month. "Again,we had planned to stop at 1 month," Dr. Cella said, "butwe had to go on because so many people were taking that extramonth even with severe toxicity."

In the end, about 10% of patients said they would accept severelytoxic therapy for an extra week of life, and 20% said they wouldtake the treatment with mild toxicity. "Clearly, there aremany patients who wish to be treated as aggressively as possible,"Dr. Cella said. "The implications of this for the futureof medical decision-making are considerable."