Christchurch Hospital,. Christchurch: C. Atkinson, B. Fitzharris, B. Hastie, B. Robinson, L. Skelton, C. Wynne .... Philadelphia' Lippincott-. Raven 1996; 403-12.
Annals of Oncology 8: 277-283, 1997. © 1997 Kluwer Academic Publishers. Printed m the Netherlands
Original article Improved quality of life with megestrol acetate in patients with endocrineinsensitive advanced cancer: A randomised placebo-controlled trial E. Beller,1 M. Tattersall,2 T. Lumley,1 J. Levi,3 D. Dalley,4 I. Olver,5 J. Page,6 E. Abdi,5 C.Wynne,7 M. Friedlander,8 D. Boadle,9 H. Wheeler,3 S. Margrie 1 & R. J. Simes1 on behalf of the Australasian Megestrol Acetate Cooperative Study Group 'NHMRC Clinical Trials Centre, University of Sydney; 2 Department Cancer Medicine, University of Sydney; 3 Royal North Shore Hospital; St. Vincents Hospital, Sydney; Royal Adelaide Hospital, Adelaide; 6Concord Hospital, Sydney; "'Chnstchurch Hospital, Chnstchurch, ^Prince of Wales Hospital, Sydney; 9 WP Holman Clinic, Launceston, Australia * Seepage 283 for list of contributors to this study
A
Summary Purpose: To investigate the effect of two doses of megestrol acetate (MA) compared with placebo on quality of life (QoL) and nutritional status (NS) in patients with advanced endocrine-insensitive cancer. Patients and methods: Two hundred forty patients were randomised to double-blind MA 480 mg/day, MA 160 mg/ day, or matching placebo for 12 weeks. Nutritional status (including weight, skinfold thickness and midarm circumference) and QoL (using 6 linear analogue self-assessment (LASA) scales) were assessed at randomisation and after four, eight and 12 weeks. A QoL ranking incorporating QoL and death was also used ranging from 1 = dead to 5 = much better QoL. Results: One hundred seventy-four patients were assessable at week four, 136 at week eight and 103 patients at week 12. Patients receiving MA reported substantially better appetite (P = 0.001), mood (P = 0.001) and overall quality of life (P < 0.001), and possibly less nausea and vomiting (P = 0.08)
Introduction
Constitutional symptoms are common in patients with advanced cancer. These include lethargy, anorexia, nausea and weakness. Weight loss is an important prognostic factor for many cancers, predicting a short survival, a low probability of response to anti-tumour treatments and increased toxicity after these therapies. The causes of constitutional symptoms in advanced cancer are not known. Synthetic progestagens, namely megestrol acetate and medroxyprogesterone, are used widely in the treatment of patients with metastatic breast cancer [1, 2]. Increased appetite and weight gain, independent of tumour response, are frequent in patients taking these drugs, and it was suggested that these side effects may be beneficial in some patients [3, 4]. The weight gain does not appear simply to be due to fluid retention, suggest-
than patients receiving placebo, based on a test for trend. A larger benefit was seen with the higher dose which (unlike the lower dose) was significantly better in pairwise comparisons with placebo for appetite, mood and overall QoL (each P ^ 0.001). Despite some missing data on QoL scores, QoL ranking was available on 227 (95%) of patients with significantly higher QoL ranking associated with MA (P = 0.002). Improvements in QoL occurred early within four weeks and were sustained. No statistically significant differences were observed in NS measurements, including weight (P = 0.29). Side effects of therapy were minor and did not differ significantly across treatments. Conclusion: Megestrol acetate given at 480 mg/day is useful palliation in patients with endocrine-insensitive advanced cancer. It improves appetite, mood and overall quality of life in these patients, although not through a direct effect on nutritional status. Key words: cachexia, megestrol acetate, quality of life
ing that megestrol acetate may play a useful role in the care of patients with advanced cancer. Several controlled trials of megestrol acetate have been undertaken in cancer patients, assessing the effect on appetite and weight gain [5-9]. In the study reported by Loprinzi et al. [6], patients were randomised to receive megestrol acetate at 800 mg/day or placebo. Patients receiving megestrol acetate had improved appetite and food intake compared with the placebo group. A crossover design study by Bruera [5], using a dose of 480 mg/day also observed that the effect of megestrol acetate was primarily on appetite and caloric intake, but also on weight, triceps skinfold thickness and calf circumference, suggesting an improvement in nutritional status. A further study by Tchekmedyian [7] used 1600 mg/ day of megestrol acetate, and demonstrated a significant effect on appetite. Another study [9] used a dose of 240 mg/day and reported significant appetite stimulation,
278 with more patients in the megestrol acetate group, (compared to placebo), gaining at least 2 kg or at least 5 kg in weight. Megestrol acetate has also been proposed as a treatment for cachexia associated with human immunodeficiency virus (HIV) infection. Recently, two randomised trials of megestrol acetate in patients with AIDS and cachexia have been reported. The study by Von Roenn [10] addressed dose-response by randomly assigning patients to receive placebo or three different dose levels of megestrol acetate (100 mg, 400 mg and 800 mg daily) for 12 weeks. The study by Oster et al. compared placebo with megestrol acetate 800 mg daily. Both trials showed that 800 mg per day megestrol acetate increased weight and daily caloric intake, and body composition measurements showed that weight gain was not related to water retention, but to increased levels of stored fat and to some extent increased lean body mass. Patient assessment of well-being was monitored by nine linear analogue scale questions investigating weight, altered appearance, appetite, food intake, enjoyment of food, perception of whether treatment is working or beneficial, and overall quality of life. We investigated whether megestrol acetate effects on appetite and weight would translate into measurable benefits for cancer patients' overall quality of life (QoL), and hence our study was designed to have quality of life as the primary outcome. Patients and methods Patient selection Patients with advanced endocrine-insensitive cancer, who had a body weight at least 5% below ideal, or unintentional loss of at least 5% of usual body weight were eligible for the study. The study was approved by the Institutional Ethics Committees of the participating hospitals, and patients gave written informed consent prior to study entry. Patients were excluded if they were under 18 years of age, had physical or functional obstruction to food intake or impaired digestive/absorptive function, were pregnant, were receiving concurrent corticosteroid treatment, were unable to complete quality of life forms, had endocrine-sensitive malignant disease (i.e., of breast, prostate, uterus), had a life expectancy of less than two months, or were diabetic.
Randomisation Patients were stratified by institution, and by whether they were receiving any anti-tumour treatment. Patients were entered onto the study and randomised by telephone through a central office. Eligibility was checked during this telephone call, prior to randomisation Two hundred forty patients were randomised to one of three treatment groups: (1) megestrol acetate 480 mg per day (higher dose group), (2) megestrol acetate 160 mg per day (lower dose group), (3) placebo control. The study was conducted in a double-blind fashion.
matching placebo in the lower dose group, and 3 x placebo in the placebo group. Four weeks' supply of tablets was distributed to patients at randomisation and at clinic visits four and eight weeks after randomisation, for a total treatment duration of 12 weeks.
Outcome assessment At baseline and at four, eight and 12 weeks after randomisation, patients were asked to complete a quality of life questionnaire incorporating the five linear analogue self-assessment (LASA) scales of Priestman & Baum [11], and a LASA uniscale of overall quality of life from Coates et al. [12]. In addition, the clinician completed the Spitzer QL-Index quality of life instrument [13] about the patient. Nutritional status was also measured at these time intervals, comprising weight, mid-arm circumference, triceps skinfold thickness, serum albumin and creatinine levels, haemoglobin and lymphocyte count and degree of oedema present. At baseline and 12 weeks, serum folate and prealbumin were also measured. Participating doctors and study nurses were given detailed instructions regarding measurement of mid-arm circumference and triceps skinfold thickness. Callipers were supplied for the triceps skinfold thickness measurements. Where possible, a single observer conducted patient measurements at all assessment times. Information about known side-effects of megestrol acetate, namely oedema, thrombosis, and dyspnoea which may result from pulmonary emboli, was collected at each clinic visit and graded on a scale of 0 = none, 1 = mild, 2 = moderate, 3 = severe, 4 = life-threatening
Statistical considerations The protocol was designed to ensure that the study would have a power of 80% at the 5% significance level to detect a difference of 0.45 standard deviations in the overall LASA QoL score or the Spitzer QL-Index. This is the level of improvement seen in quality of life scores for the ANZ Breast Cancer Trials Group's ANZ8101 trial [14], and was thought to represent a plausible and worthwhile improvement in quality of life. Therefore, the sample size of 240 patients was chosen.
Statistical analysis There were three primary outcomes: a combined measure of quality of life calculated from the LASA scale responses of the patients, the Spitzer QL-Index and a combined nutritional status score calculated from the best nutritional predictors of quality of life at randomisation and over the first four weeks of treatment. Details of the construction of the two combination measures are discussed in the following sections. Secondary outcomes included each of the separate quality of life and nutritional status measurements, and survival time. The main analyses used the Generalised Estimating Equation (GEE) method for repeated measurements [15] to compare the changes in quality of life and nutritional status from baseline to weeks four, eight and 12. The GEE approach takes into account the correlation between different measurements on the same subject and so provides more precise and accurate estimates of the treatment effect. Survival times were analysed using the log-rank test and Cox regression. All these analyses were performed using the SPIDA statistical package [16]. Two-sided 5% level tests were used throughout, and no adjustment was made for multiple comparisons.
Quality of life measures Treatment All patients received three tablets per day, as 3 x 160 mg megestrol acetate in the higher dose group. 1 x 160 mg megestrol acetate and 2 x
The study had multiple outcome measures for two of the main outcomes, quality of life and nutrition. The LASA quality of life questionnaire asked patients about five factors related to their quality of life. (1) physical well-being, (2) mood, (3) pain, (4) nausea and vomiting, and
279 (5) appetite. This questionnaire also asked for a rating of overall quality of life. This uniscale measurement of overall quality of life has the advantage that it measures what the patient considers most important, and so is closer to a true utility measurement. As the other LASA scales are correlated with the uniscale, it is possible to use information from them to obtain a more precise estimate of overall quality of life than is given by the uniscale measure alone. The linear combination of the five LASA scales which best predicted the uniscale measurement was calculated by linear regression. This linear combination measure was then combined with the uniscale itself, using a weighted average, with weights inversely proportional to their mean square errors [17]. When these weights were derived, they were found to match very closely similar estimates from the ANZ breast cancer trial referred to earlier. In order to assess the effect of missing QoL data on conclusions, a quality of life ranking was obtained for almost all patients based on the patient's clinical status whenever quality of life scores were unavailable. For patients with LASA scores, changes in the combined quality of life score were categorised as worse (below -5), same (-5 to 5), better (5 to 15) or much better (more than 15) with allocated scores of 2, 3, 4 and 5, respectively. A similar ranking was applied to the patient's QL-index when this but not the LASA score was available. Patients who had died were given a score of 1. Patients who left the study due to deteriorating condition or drug intolerance or toxicity were scored as 2 (worse). This enabled the derivation of a QoL ranking for 95% of patients, and can be regarded as a form of sensitivity analysis for handling missing QoL data.
Nutritional status measures Measures of nutritional status collected in the trial were weight, midarm circumference, triceps skin fold thickness, (from which mid-arm fat and muscle areas were derived [18]), absence of oedema, serum albumin, prealbumin, creatinine, folate, haemoglobin and lymphocyte counts. In view of these multiple measures, a single summary measure of nutritional status was developed which was most closely associated with changes in quality of life. In a multivariate linear regression analysis the best predictors of changes in quality of life were weight, mid-arm fat area and serum albumin. Hence, a weighted sum of these three variables was used to give primary nutritional status outcome. This analysis was performed without knowledge of treatment allocation or the main treatment results of the trial.
Results Patient accrual Two hundred forty patients were randomised into the study from 15 institutions (range 1-44 patients per institution), with 79 in the placebo group, 80 in the lower dose group and 81 in the higher dose group. Of these, 10 had missing baseline measurements that precluded the calculation of a nutritional status score, leaving 230 for whom the full baseline evaluation was possible (Table 1). Withdrawals from randomised treatment, (other than death whilst still on treatment), included drug intolerance or toxicity in 28 cases, deterioration in patient's condition not attributed to study treatment in 36 cases, and refusal by patient to continue on the study in 18 cases. The reason for the remainder of withdrawals from the study is unknown. The proportion of incomplete data and withdrawals from treatment was very similar for the three treatment groups.
Table 1. Availability of data (number of patients with data at each assessment time).
Alive Alive with partial data 1 Alive with full data
Baseline
4 weeks
8 weeks
12 weeks
240 240 230
201 174 148
179 136 112
147 103 83
* At least one of the quality of life or nutritional status measures missing.
Table 2. Baseline characteristics. Placebo (n = 79)
Age 12 70 Gender Female 25 Performance status (ECOG grades) 0 12 1 32 2 22 11 3 Concurrent antitumour therapy* None 32 Chemotherapy 45 Other 3 Baseline weight 1OSS (% loss from usual weight) 15% 38 Tumour type GI 32 Lung/pleura 18 Head & neck 5 4 Pancreas Lymphoma/ leukemia 3 Other 17
Lower Higher dose dose (n = 80) (n = 81)
Total (%) (n = 240)
7 52 21
9 49 23
28(12%) 152(63%) 60 (25%)
28
28
81 (34%)
P-value (y_2)
0.60
0.89 0.82 8 32 28 11
9 38 24 8
29(12%) 102 (43%) 74(31%) 30(13%) 0.40
33
32
97 (40%)
44 6
41 10
130(54%) 19(8%) 0.54
17 20 43
19 17 45
52 (23%) 62 (25%) 126(52%)
32 13 4 3
31 17 9 4
95 (39%) 49 (20%) 18(8%) 11 (5%)
6 22
3 17
12(5%) 56 (23%)
0.84
* Some patients were receiving more than one modality.
Pretreatment patient characteristics The three treatment groups are comparable on all measured baseline characteristics (Table 2). Of the 240 patients randomised, 143 (60%) were receiving concurrent anti-tumour therapy, mostly chemotherapy. The tumour sites were predominantly gastrointestinal (39%) and lung/pleura (20%). Thirty-four percent of patients were female, and 77% had an unintentional weight loss of at least 10% of usual weight, with 52% having lost more than 15% of usual weight.
280 Quality of life results
Changes in appetite lime profiles and 90% confidence half-intervals
The combined quality of life measure showed a significant and substantial improvement in the high dose group compared to placebo (P < 0.001), and a smaller, non-significant improvement was seen in the lower dose group (Table 3). The changes from baseline to week 4, to week 8 and to week 12 were similar, with no evidence of a time effect (Figure 1). Similar results and patterns of improvement were observed in the LASA scales of appetite (P = 0.001), mood (P = 0.001) and overall quality of life (P < 0.001) (Figures 2-4). Both megestrol acetate treatment groups had a non-significant improvement on the nausea and vomiting rating. The largest effects were on appetite, as has previously been reported by others. The results were confirmed using the derived QoL
50
480mg/day
160mp/day placebo
baseline
Week 4
Week 8
Week 12
Figure 2. Time profile of changes in appetite LASA scale measure, with 90"'.. confidence intervals.
Changes in mood Table 3. Quality of life results. Average difference between baseline and subsequent weeks. Measure
Placebo
LASA scores" Physical well-being Mood Pain Nausea/vomiting Appetite Overall QoL Combined QoL measure b QL-Index c QoLranking d
Lower dose
Higher dose
P-value (trend)
time profiles and 90% confidence half-intervals 15
T
10
r
T
5 r
A
5.8 -4.1 -5.3 -1.4 9.7 -2.7 -2.1 -0.15 2.37
6.5 0.4 -6.9 8.7 17.0 28 2.4 -0 27 2.66
13.9 10.2 19 7.2 31.3 13.1 12.3 0.26 3.05
0.13 0.001 0.13 0.08 0.001 < 0.001 < 0.001 0.30 0.002
" Linear analogue self-assessment scales (LASA). Measured on 100 point scale. b Weighted combination of individual LASA scores Weights were derived from regression analysis of individual scores on overall QoL. Measured on 100 point scale. c Spitzer QL-Index instrument. Measured on 15 point scale. d QoL ranking categorised for 95% of patients: 1 - death; 2 - QoL worse; 3 - QoL same; 4 - QoL better; 5 - QoL much better. Measured on 5 point scale.
480mg/day 160mg/day
-5
1
10 15
p=0.001 (trend with dose)
' - - - ,i
placebo
I
20
baseline
Week 4
WeekS
Week 12
Figure 3. Time profile of changes in mood LASA scale measure, with 90".,confidence intervals.
Changes in overall quality of life time profiles and 90% confidence half-intervals 20
10 480mg/day 160mg/day
Changes in combined QoL estimate
i
time profiles and 90% confidence half-intervals -10
J
p < 0.001 (trend with dose) 480mg/day -20
baseline 160mg/day
placebo
baseline
placebo I
Wee-
Figure 1. Time profile of changes in the combined quality of life measure, with 90% confidence intervals.
Week 4
Weeks
Week 12
Figure 4. Time profile of changes in overall quality of life uniscale LASA measure, with 90"" confidence intervals.
rank, referred to earlier. This also demonstrated a significant treatment effect for higher dose megestrol acetate compared with placebo (P - 0.002) (Table 3). Changes in Spitzer's QL-Index did not differ with treatment. This may be because the dimensions measured by the QL-Index are not strongly related to the
281 areas in which the LASA scales showed an improvement. Table 3 summarises the quality of life analyses. At randomisation, patients were stratified into two groups based whether or not they were receiving antitumour treatment. A separate analysis for each group shows that the effect of megestrol acetate on overall quality of life was substantially greater for those receiving concurrent anti-tumour treatment, although the difference between stratification groups is not statistically significant. Nutritional status results There was no significant difference in weight change between treatment groups (P = 0.29). However, of the 32 patients (20%) with the most weight gain (greater than 3.65 kg), 15 were in the higher dose group and 9 in the lower dose group. A repeated measures analysis of the combined nutritional status score showed no statistically significant difference between the three treatment groups (Table 4). In a secondary analysis, the changes in the individual measures of nutritional status were analysed by the same method without detecting any differences between groups (Table 4). Survival There was no difference in survival between the three treatment groups, either over the 12-week study period, or subsequently. However, the trial was not designed with sufficient power to detect clinically meaningful differences in survival. Survival plots are shown in Figure 5.
Table 4. Nutritional status results. Average difference between baseline and subsequent weeks. Measure
Weight (kg) Combined nutritional status score Triceps skinfold (mm) Mid-arm circumference (cm) Mid-arm fat area (cm 2 ) Mid-arm muscle area (cm2) Haemoglobin (g/dl) Lymphocyte count (10 9 /l) Creatine/height ratio (umol/(cml)) Albumin (g/1) Folate (ug/1) Prealbumin (mg/1)
Pvalue (trend)
bo
Lower dose
Higher dose
-0.53
-0.66
0.15
0.29
-1.88 -0.28
-0.41 -0.70
-2.46 0.15
0.72 0.45
-0.62 -1.34
-0.30 24.34
-0.31 -3.24
0.35 0.17
0.81 -0.54 0.03
-8.68 -0.52 0.04
0 26 -0.33 0.18
0.91 0.38 0.17
-0.010 -0.73 0.59 -7.61
0 037 -0.33 0.78 4.00
0 048 -0.92 -1.92 13.34
0.22 0.83 0.20 0.24
Place-
severe or life-threatening dyspnoea, four in the higher dose group, six in the lower dose group and one in the placebo group. Five of these 11 patients had primary lung cancer or lung metastases from other primary sites. One of the patients with severe dyspnoea also had a pulmonary embolus. This patient had metastatic cancer from an unknown primary site. Another patient with moderate dyspnoea had a suspected pulmonary embolus, not confirmed at later autopsy. Both patients with pulmonary emboli were in the lower dose megestrol acetate group. Four patients had severe oedema, all in the lower dose megestrol acetate group.
Side effects
Discussion
There was no statistically significant difference between treatment groups in occurrence or severity of adverse effects. Clinicians were not asked to report whether adverse effects were related to the study drug, so these results may include reported adverse effects of concomitant medications also. However, 11 patients experienced
We found significant improvements in patient-rated overall quality of life in those who received megestrol acetate when compared to placebo. This improvement was seen within four weeks after commencing treatment, and was sustained for the twelve weeks of treatment. The patient-rated LASA scales used to assess quality of life, examined physical well-being, mood, pain, nausea and vomiting, and appetite. Megestrol acetate significantly improved mood and appetite, and there was a trend for improvement of nausea and vomiting. However, physician-rated quality of life, using a different instrument, was not significantly different between the three treatment groups. The measurement instrument used by physicians investigates five quality of life domains: activity, daily living, health, support and outlook, and these are not closely related to the specific aspects of quality of life that patients rated as significantly improved by megestrol acetate. Together these results indicate the contribution of different quality of life dimensions to the perceived well-being of patients entered in the trial, and underline the importance of having a patient-rated overall measure of QoL.
% surviving
— Placebo — 160mg?day — 480mg.'day
0
20
40
60
Weeks since randomization Figure 5 Survival curves for the three treatment groups (KaplanMeier estimates).
282 A proportion of patients in this trial had missing quality of life data particularly with longer follow-up. Other trials of megestrol acetate have had similar problems with missing data. Loprinzi et al. [8] reported 50%-60% response rates (for patients still alive) for some of their quality of life questions. Tchekmedyian et al. [7] reported about 25% of their subjects inevaluable after one month. Since withdrawals may reflect patients in poorer health, an additional analysis was undertaken using a QoL ranking based on the patient's clinical status when quality of life scores were unavailable. This analysis based on 95% of the patient population confirmed a benefit for megestrol acetate and suggests that missing QoL data has not biased the trial results. Our results are similar to the report of Downer et al. [19] in that they also noted improvement in appetite though not in mood in anorexic patients who received another progestational agent, medroxyprogesterone acetate (MPA) 100 mg tds, for three to six weeks. They also observed a significant increase in serum thyroid binding, pre-albumin and retinol binding protein in the treated group compared to control, but like us they noted no change in weight with treatment. Similarly Simons et al. [20] have reported an improvement in appetite, and a trend toward a reduction in nausea and vomiting with MPA treatment. They also observed that patients who are weight-losing but not yet cachectic may benefit from MPA treatment. Several previous studies of synthetic progestagens in patients with AIDS or cancer-related cachexia have reported significant weight gain, as had originally been noted with progestational treatment of breast cancer patients, but we observed no differences in weight or nutritional status in the three randomised groups. Whilst some earlier studies used similar megestrol acetate doses to ours, and reported weight gain, most studies in which weight gain has been observed used substantially higher doses, and weight gain was more common at the highest dose investigated. Higher megestrol acetate doses might have increased weight in our patients but at an increased drug cost and perhaps side effects. We had chosen to evaluate both 160 mg and 480 mg doses to see if there was perhaps a balance between benefit and cost of the drug. It can be argued that weight gain itself is not as important a treatment endpoint as the improvement in appetite, mood and overall well-being which we noted in patients receiving 480 mg megestrol acetate per day. At the lower megestrol acetate dose of 160 mg per day used in our study, we observed a non-significant improvement in quality of life measurements, suggesting that there may be a dose-response relationship even for these more relevant quality of life measures. The association that others have reported between megestrol-related increase in weight and patient-rated quality of life is likely due to the drug effects being interrelated. Future studies in AIDS and cancer-related cachexia should investigate the dose of progestational agent which achieves the maximum improvement in quality of life with the lowest side effects and cost.
It is notable that corticosteroids used in cancer treatment and palliation are commonly observed to stimulate appetite and reduce nausea and vomiting which may accompany cancer treatment. We found that megestrol acetate had precisely these effects. Is the beneficial effect we have observed due to a corticosteroid like action of megestrol acetate? A randomised trial in cancer-related cachexia comparing megestrol acetate with prednisone was unfortunately abandoned due to lack of accrual after 40 patients (A. Goldhirsch, personal communication). Suppression of baseline cortisol levels have been observed in some patients taking progestational agents, suggesting an effect on the pituitary-adrenal axis. Our observation that the effect of megestrol acetate on overall quality of life was greater for those receiving concurrent antitumour treatment may prompt clinicians to include progestational agents during rather than after antitumour treatment has been abandoned. In keeping with this observation Simons et al. [20] in a subgroup analysis suggested that patients who are weight-losing but not yet cachectic may benefit most from treatment with progestational agents. The antinausea/antiemetic effect of corticosteroid has been demonstrated in randomised trials of antiemetic regimens given with cytotoxic chemotherapy. It is possible that a similar benefit and mechanism of megestrol acetate is operating in cachectic patients receiving active cancer treatment. The absence of significant treatment-related toxicity in our study is reassuring, though the possibility of a drug-related contribution to dyspnoea is not excluded. The lack of an effect of treatment on survival is not surprising, though some might have feared a reduction in survival time due to 'feeding' of the tumour which increased caloric intake might have induced. The goal of therapy in these cachectic patients is to enhance appetite, mood and well-being, and perhaps reduce weight loss. The quality of life impact of weight gain per se is not established nor plausible. AIDS-related cachexia may be a special case since some studies of patient quality of life in this setting have monitored appearance and found this to be associated with other more traditional quality of life domains. In cancer cachexia, patient appearance has not been monitored as a quality of life domain, and lack of weight gain may be less important when appetite and well-being are enhanced. We also noted that the benefit of megestrol acetate was evident at four weeks, and presumably earlier. If this effect is not mediated through improvement of nutritional status, then future studies should consider earlier evaluation of response. Additional studies of synthetic progestagens and corticosteroids on the quality of life of patients with advanced cancer who are losing weight are justified. These should examine when to commence treatment, the dose, and the advantages of progestagens over corticosteroids.
283 * Contributors to this study Austin Hospital, Melbourne. K. Hynes, L. Kelsey, J. McKendrick, G Moystyn. R. Woodruff. Box Hill Hospital, Melbourne. J. Chirgwin, R. Chuang, M. Dooley, D. Goldstein, M. Leydon. Christchurch Hospital, Christchurch: C. Atkinson, B. Fitzharris, B. Hastie, B. Robinson, L. Skelton, C. Wynne Concord Hospital, Sydney: R. Aroney, E. Moylan. J. Oakley, J. Page, L. Truong, M. Wright. Dunedin Hospital, Dunedin: R. Carlson, J. Gonin, D Perez, B Rae. Heidelberg Repatriation Hospital, Melbourne: W. Cosolo, J. Laird, J. Siderov, J. Zalcberg, A. Zimet. Monash Medical Centre & Prince Henry's Hospital, Melbourne: Y. Allinson, E. Athan, P. Briggs, G. Brodie, T. Cropley, S. Curza, J. Griffiths, R. Stanley NHMRC Clinical Trials Centre, Sydney (coordinating centre): E. Beller, J. Fabri, L. Helby, A. Hill,T. Lumley, S. Margrie, M. Mulders, R. Philip, J Simes, C. Stone, L. Wilcox. Peter MacCallum Cancer Institute, Melbourne: H. Baxter, J. Bingham, J. Bishop, W Mercer, R. Murray. Prince of Wales Hospital, Sydney J. Campbell, B. Fardell, M. Friedlander, C. Lewis, M. Ng, A. Prendergast. Princess Alexandra Hospital, Brisbane: G. Beadle, G. Pinna, D. Thompson, E. Walpole, C. Wingate. Royal Adelaide Hospital, Adelaide: E. Abdi, J. Goulding, I. Olver, A. Robertson, E. Zerella. Royal North Shore Hospital, Sydney: D. Bell, M. Bocking, L. Hayden, J. Levi, S. McCowatt, D. Paull, C Stewart, N. Tadros, H. Wheeler. Royal Prince Alfred Hospital, Sydney. A. Coates, D. Cunningham, J. Grygiel, E. McNeil, N. Sieger, J. Simes, A. Sullivan, M. Tattersall, N. Teriana. St. Vincents Hospital, Sydney: D & D DalleyT. Melocco. WP Holman Clinic, Launceston: D. Boadle, G. Wighton.
References 1. Hedley D, Dalgleish A, Raghavan D et al. Advanced breast cancer: Response to high dose oral medroxyprogesterone acetate Aust N Z J Med 1984; 14: 251-4 2. Schacter L, Rozencweig M, Canetta R et al. Megestrol acetate: Clinical experience. Cancer Treat Rev 1989; 16. 49-63. 3. Tchekmedyian NS, Tait N, Moody M et al. Appetite stimulation with megestrol acetate in cachectic cancer patients. Semin Oncol 1986; 13 (Suppl 4): 37-43. 4. Aisner J, Tchekmedyian N, Parnes H, Novak M. Studies of higher dose megestrol acetate: Potential applications in cachexia. Semin Oncol 1988; 15 (Suppl 1)- 68-75. 5. Bruera E, Macmillan K., Kuehn N et al. A controlled trial of megestrol acetate on appetite, caloric intake, nutritional status, and other symptoms in patients with advanced cancer. Cancer 1990; 66. 1279—82. 6. Loprinzi C, Ellison N, Schaid D et al. Controlled trial of megestrol acetate for the treatment of cancer anorexia and cachexia. J Natl Cancer Inst 1990; 82: 1127-32. 7. Tchekmedyian NS, Hickman M, Siau J et al. Megestrol acetate in cancer anorexia and weight loss. Cancer 1992; 69: 1269-74.
8. Loprinzi C, Johnson P, Jensen M. Megestrol acetate for anorexia and cachexia Oncology 1992; 49 (Suppl 2): 46-9. 9. Feliu J, Gonzalez-Baron M, Berrocal A et al. Treatment of cancer anorexia with megestrol acetate: Which is the optimal dose? J Natl Cancer Inst 1991; 83. 449-50. 10. Von Roenn JH, Armstrong D, Kotler DP et al. Megestrol acetate in patients with AIDS-related cachexia. Ann Intern Med 1994; 121: 393-9. 11. Priestman TJ, Baum M. Evaluation of quality of life in patients receiving treatment for advanced breast cancer. Lancet 1976; 1. 899-901. 12. Coates A, Glasziou P, McNeil D. On the receiving end - III. Measurement of quality of life during cancer chemotherapy. Ann Oncol 1990; 1: 213-7. 13. Spitzer WO, Dobson AJ, Hall J et al. Measuring the quality of life of cancer patients. A concise QL-Index for use by physicians. J Chronic Dis 1981, 34: 585-97. 14. Coates A, Gebski V, Bishop JF et al. Improving the quality of life during chemotherapy for advanced cancer: A comparison of intermittent and continuous treatment strategies. N Engl J Med 1987; 317: 1490-95. 15. Zeger S, Liang K-Y. Longitudinal data analysis for discrete and continuous outcomes. Biometrics 1986; 42: 121-30. 16. SPIDA (Statistical Package for Interactive Data Analysis) User Manual. Statistical Computing Laboratory Pty Ltd, Australia 1992. 17. Martin AJ, Lumley TS, Simes RJ. Incorporating trade-offs in quality of life assessment. In Spilker B (ed): Quality of Life and Pharmacoeconomics in Clinical Trials. Philadelphia' LippincottRaven 1996; 403-12. 18. Rudman D, Bleier JC Assessment of nutritional status. In Petersdorf RG, Adams R, Braunwald E et al (eds): Principles of Internal Medicine. Japan: McGraw-Hill 1983; 433-6. 19. Downer S, Joel S, AJlbright D et al. A double-blind placebocontrolled trial of medroxyprogesterone acetate in cancer cachexia. Br J Cancer 1993; 67. 1102-5. 20. Simons J-P FHA, Aaronson NK, Vantsteenkiste JF et al. Effects of medroxyprogesterone acetate on appetite, weight and quality of life in advanced-stage, non-hormone sensitive cancer. A placebo-controlled multicentre study. J Clin Oncol 1996; 14: 1077-84 Received 13 September 1996, accepted 16 January 1997.
Correspondence to • John Simes, MD NHMRC Clinical Trials Centre University of Sydney Locked Bag 77 Camperdown NSW 2050 Australia
