3803

Quality of Life in Prostatic Cancer Patients F. Calais da Silva, M.D.

As a first effort to introduce quality-of-life assessment in prostatic cancer clinical trials, the European Organiza- tion for Research and Treatment of Cancer Genitouri- nary Group, in cooperation with European Organization for Research and Treatment Quality of Life Group, initi- ated protocol 30853, coordinated by Louis Denis. This protocol compared the efficacy of treatment with or- chiectomy alone to that with Zoladex (Zeneca Pharma- ceuticals, Alderley Macclesfield, Cheshire, UK) plus flu- tamide in previously untreated patients with metastatic cancer. The use of patient-administered quality-of-life questionnaires was optional, and of 327 patients, only 22% had pretreatment assessments. This trial revealed many clinician’s considerable reluctance to perform qual- ity of life research, partly because of feasibility problems and partly because of doctors’ doubts about the value of such efforts. Psychologic distress, fatigue, issues of social and family life, and pain were found to be the most im- portant concerns on a subjective basis, and this finding was confirmed by objective parameters. There was a dis- crepancy between doctors’ evaluations and patients’ opinions about subjective morbidity, namely sexual sta- tus and pain, Quality of life assessment should become a mandatory part of clinical trials in prostate cancer. Cancer 1993: 723803-6.

Key words: quality of life, prostate cancer, question- naires assessment.

During recent years there has been an increased interest in quality of life investigation. Perhaps the most impor- tant reasons for the failure to incorporate quality of life evaluations into clinical trials is the lack of clarity in definitions of the concepts.

At one level, introduction of the term ”quality of life” into the medical lexicon has served an important ideologic function. It points our collective attention to- ward a broader set of issues surrounding the human

Presented at the Third International Workshop on Randomized Trials on Maximal Androgen Blockade in M1 Prostate Cancer Pa- tients, Paris, France, June 19, 1992.

From the Hospital do Desterro, Lisbon, Portugal. Accepted for publication August 13, 1993.

costs of disease and medical treatment. In this sense, the term ”quality of life” serves a function similar to that of the World Health Organization definition of health: ”a state of complete physical, mental and social well-being, and not merely the absence of disease or infirmity”.

It generally is recognized that the World Health Or- ganization definition of health is intended to serve pri- marily a theoretical function and is not meant to be translated directly into research applications. This has not, however, always been the case with quality-of-life research. Rather, there is a tendency to reify the quality of life concept. This had lead to attempts at global mea- surement of quality of life in cancer research. For exam- ple, Gough et al.’ suggested that a single, direct ques- tion-How would you rate your quality of life today?-represents a valid and reliable approach to as- sessing the well-being of cancer patients. The obvious difficulty with such an approach is that it provides little or no information that can be used to improve the clini- cal management of patients. How are we to interpret a patient reporting a low overall quality of life? Does it imply intolerable symptom levels, limitations in func- tional capacity, psychological morbidity, a disrupted social life, or a combination of these factors? Unless the quality-of-life concept is divided into more easily ana- lyzed units, it is unlikely that it will achieve widespread acceptance as a legitimate, let alone routine, part of clinical research.

A second, related barrier to the introduction of qual- ity of life parameters in clinical trials is the lack of con- sensus on how it should be measured. The physician interested in assessing the psychosocial impact of both routine medical care and experimental treatments is confronted with a confusing array of measurement tools. At one end of the spectrum are extensive and time-consuming interview protocols that exceed the practical limitations present in the typical clinical set- ting. At the other end are more concise patient self-re- port questionnaires that, while representing a less cum- bersome approach to data collection, often leave unan- swered the critical questions of instrument reliability and validity.

3804 CANCER Supplement December 15, 2993, Volume 72, No. 12

Table 1. Pretreatment Assessment Doctors Patients

Pain 46 52 Symptoms 29 27 Decreased performance status 15 41 Potent 29 57

Similarly, psychosocial research designs often are viewed by those responsible for conducting clinical trials as overly burdensome to both themselves and their patients. Such a perceived burden may be due, in part, to a lack of familiarity with the basic nature of requisite research and intent of assessing psychoso- cia1 outcomes. However, it also reflects a degree of naivitk on the part of social scientists with regard to what is feasible in a clinical context. Frequently, data collection schemes call for the frequent completion of lengthy questionnaires in the midst of the rush en- demic in out-patient clinics. Although such proce- dures may be optimal in an abstract sense (i.e., they potentially can yield a maximum amount of data), they do not address the practical constraints that exist clinical practice.

Data collection represents an additional burden to the patient and the clinician involved in the clinical trial. Patients may need assistance, especially when in- troduced to the questionnaires for the first time. There- fore, questionnaires should be as short as possible, and the frequency of assessments should be as low as pos- sible.

As a first effort to introduce quality-of-life assess- ment into clinical trials for prostatic cancer, the Euro- pean Organization for Research and Treatment of Cancer (EORTC) Genitourinary Group, in cooperation with EORTC Study Group on Quality of Life, designed a study on the quality of life of patients enrolled in the EORTC 30853 protocol on metastatic prostate cancer. EORTC 30853, coordinated by Louis Denis, is a ran- domized phase I11 trial comparing the therapeutic effect of orchiectomy with that of LHRH-analog depot prepa- ration plus Flutamide in patients with metastatic pros- tatic cancer.* Although the primary end points of the trial are incidence and duration of response, time to progression and overall survival also are being studied.

Quality of life was an optional aspect of the overall analysis determining the relative impact of the two therapies on the daily lives of the patients. Toward this end, a 30-item questionnaire was developed to assess: (1) personal functioning (i.e., a self-report analog of physician-rated performance status); (2) social role

functioning; (3) physical symptoms of prostatic cancer (i.e., dysuria, frequency of urination, hematuria, meta- static pain); (4) fatigue and malaise; (5) sleep distur- bance; (6) psychologic distress; (7) sexual dysfunction; and (8) disruption of social life (see Appendix). The ma- jority of the items and scales composing the question- naire have been used in previous EORTC trials and have established levels of validity and reliabilit~.~

Patients and Methods

Twenty-two institutions entered 327 patients untreated with metastatic prostatic cancer into EORTC protocol 30853. In the collaborating centers, the patients were asked at pretreatment and at each control period (3 months) to fill in a questionnaire on quality of life as already described.

In spite of several shortcomings of the question- naire concerning quality-of-life assessment, some inter- esting observations were made between patients’ and physicians’ assessments of parameters thought to be significant for the quality of life of a prostatic cancer patient.

Results

Doctors’ assessments at pretreatment of the 76 patients in four categories-pain, urologic symptoms, perfor- mance status, and sexual status (potency)-were ana- lyzed. Forty-six patients reported pain (score-l: no pain; 2: a little pain; 3: quite a bit of pain; 4: a lot of pain), 29 reported urologic symptoms, 15 had de- creased performance status, and 29 were potent (Table 1). These findings were in contrast with those based on the patients’ questionnaires, which indicated that 41 patients had decreased performance status and 57 were potent.

After 6 months, follow up with doctors’ assess- ments and quality of life questionnaires was completed for 48 patients. At this point, according to patients’ questionnaires, 25 patients had no pain, 6 claimed to be potent, 23 had decreased performance status, and 21 had no symptoms. Doctors’ assessments revealed that 31 patients had no pain, all patients were impotent, only 6 patients, surprisingly, had decreased perfor- mance status, and 31 had no symptoms (Table 2).

After 12 months, follow up with patient question- naires was completed for 29 patients. According to these questionnaires, of these patients, 17 had no pain, 5 were still potent, and 12 had decreased performance status, and 23 had no symptoms. On the other hand, doctors reported that 23 of these patients had no pain,

Quality of Life in Prostatic Cancer Patients/Calais da Silva 3805

Table 2. Follow-up, 6 Months (48 Patients)

Doctors Patients

No pain No symptoms

31 25 31 21

Decreased performance status 6 23 6 Potent -

only 4 patients had decreased performance status, and all were impotent, and 20 had no symptoms (Table 3).

The limited number of patients who answered the questionnaire makes it difficult to analyze psychologic aspects, and the questionnaire used did not include any personal questions about quality of life (e.g., "How would you rate your overall quality of life?") but in- cluded four questions about psychologic well-being (i.e., "Did you feel tense?", "Did you feel irritable?" "Did you worry?", and "Did you feel depressed?"). A score derived from the answers of these four questions was used to estimate patients' overall quality of life.

Simple correlation analysis (Pearson's correlation coefficient, R) revealed some domains that were corre- lated particularly with quality of life in a patient with prostate cancer (Tables 4, 5): Fatigue, reduced social life, and impaired sexual life played an important role in the overall psychologic well-being of previously un- treated prostate cancer patients. This analysis confirms that a patient's assessment is more useful than a doc- tor's. Correlation coefficients between psychological well-being and a doctor's assessment of performance status and the use of analgesics were not found to be statistically ~ignificant.~

Discussion

The present observations should be considered with caution. Only 23% of all eligible patients in EORTC protocol 30853 were given a pretreatment assessment. With this limited sample, guidelines can be given only to the key questions highlighted by the analysis of the questionnaires.

Table 3. Follow-up, 12 Months (29 Patients)

Doctors Patients

No pain 23 17 No symptoms 23 20 Decreased performance status 4 12

5 Potent -

Table 4. (Trial 30853) Correlation Coefficients (R) Between Overall Psychological Well-being and Other Domains of Quality of Life in Prostate Cancer

Patient's assessment

R P value

Fatigue 0.406 < 0.001 Reduced social life 0.430 < 0.001 Reduced sexual life 0.362 0.003

Reduced professional life 0.285 0.023 Pain 0.190 0.124

Urologic problems 0.280 0.021

There was a considerable amount of noncom- pliance in completion of the questionnaires. Comple- tion was optional and represented the Genitourinary Group's first attempt to assess quality of life in patients with urologic cancer. When the trial started, most uro- logists were not in the habit of systematically assessing the quality of life in their patients. Few clinicians were willing to make the regular necessary effort. They prob- ably did not feel confident that this type of assessment would add any information of value to that from clini- cal, biochemical, and radiologic examination.

Our results are consistent with those emerging from EORTC protocol 30865: showing great variations between patients and doctors evaluations of perfor- mance status and sexual status (potency). This data on performance status and potency supports the view that maybe those parameters should be evaluated by the patients6 and that information relevant to patients' qual- ity of life is not sufficiently obtained by doctors in the clinical routine. This finding may have significant thera- peutic consequences. If a patient's complaints are not recognized adequately, they often are not treated suffi- ciently. Furthermore, underestimation of symptoms by the doctor may lead to a patient's dissatisfaction with health service.

It is difficult to draw any conclusions from this lim- ited sample. However, we can suggest the following: (1) the use of self-administered questionnaires can be insti-

Table 5. (Trial 30853) Correlation Coefficients (R) Between Overall Psychological Well-Being and Other Domains of Quality of Life in Prostate Cancer

Doctor's assessment

R P value

Performance status (WHO) 0.113 0.360 Pain 0.024 0.844

3806 CANCER Supplement December 15, 1993, Volume 72, No. 12

tuted; (2) the clinician’s interest is an important condi- tion for the successful application of quality of life; (3) quality of life should be mandatory in clinical trials; and (4) the currently used questionnaires are still under in- vestigation and should be reevaluated regularly.

References

5. Do you have any trouble either taking a walk

6. Do you have any trouble bending, lifting, or

7. Are you limited in any way in doing your work

8. Does your condition keep you from working at

or climbing a few flights of stairs?

stooping?

or household jobs?

1.

2.

3.

4.

5.

6.

Gough IR, Furnival CM, Schilder LG. Eur J Clin Oncol 1983;

Denis L, Robinson M, Mahler C. Protocol for a randomized pro- spective study of the treatment of patients with metastatic pros- tatic cancer to compare the therapeutic effect of orchidectomy versus LHRH-analog depot (Zoladex) preparation supple- mented by an anti-androgen. Brussels: EORTC Genitourinary Cooperative Group, 1986. Aaronson NK. EORTC protocol 30853: development of a core quality of life questionnaire for use in cancer clinical trials. Brus- sels: EORTC Data Center, 1987. Fossa S, Aaronson NK, Kaasa S, Calais da Silva F, Suciu S, Hen- geveld MW. Quality of life: evaluation in prostate cancer pa- tients. Presented at the First International Congress of the Dutch Urological Association; 1991 Oct 9-13. Fossa S, Aaronson N, Newling D, Van Cangh PJ, Denis L, Kurth KH, et al. Quality of life treatment of hormone resistant meta- static prostatic cancer. Eur J Cancer 1989; 26(11/12):1133-6. Slevin ML, Plant H, Lynch D, Drinkwater J, Gregory WM. Who should measure quality of life the doctor or the patient? Br Cancer 1988; 57:107-12.

19~1161-5.

Appendix

Quality of Life Questionnaire

Questions 1-8 were answered either YES or NO. 1. Do you need help eating, dressing, bathing, or

2. Do you have to stay in bed or a chair for most

3. Do you have to stay indoors most or all of the

4. Do you have any trouble either walking a short

using the toilet?

of the day?

day?

distance or climbing one flight of stairs?

a job or doing household jobs? Questions 9-30 were answered NOT AT ALL, A LIT-

TLE, QUITE A BIT, or VERY MUCH 9. Did you have to urinate more frequently than

is normal for you? 10. Did you have difficulty controlling your urina-

tion? 11, Did you pass blood when you urinated? 12. Did you have pain when you urinated? 13. Did you have pain in other parts of your body? 14. How much did treatment for pain help? 15. Did you feel energetic? 16. Were you physically well? 17. Were you tired? 18. Did you need to rest? 19. Did you have trouble sleeping? 20. All in all, did you feel ill? 21. Did your condition limit your interest in sex? 22. Were you limited in your ability to have an

23. Did your condition interfere with your enjoy-

24. Did you feel tense? 25. Did you feel irritable? 26. Did you feel lonely? 27. Did you worry? 28. Did you feel depressed? 29. Has your condition interfered with family or

30. Has your medical treatment interfered with

erection?

ment of sex?

social life?

your family/social life?