Megestrol Acetate for Treatment of Anorexia-Cachexia Syndrome (Review)

Home , Fluoxymesterone, Nandrolone acetate

Megestrol acetate for treatment of anorexia-cachexia syndrome (Review)

Berenstein G, Ortiz Z

This is a reprint of a Cochrane review, prepared and maintained by The Cochrane Collaboration and published in The Cochrane Library 2005, Issue 2 http://www.thecochranelibrary.com

Megestrol acetate for treatment of anorexia-cachexia syndrome (Review) Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. TABLE OF CONTENTS HEADER...... 1 ABSTRACT ...... 1 PLAINLANGUAGESUMMARY ...... 2 BACKGROUND ...... 2 OBJECTIVES ...... 3 METHODS ...... 3 RESULTS...... 5 DISCUSSION ...... 9 AUTHORS’CONCLUSIONS ...... 9 ACKNOWLEDGEMENTS ...... 10 REFERENCES ...... 10 CHARACTERISTICSOFSTUDIES ...... 13 DATAANDANALYSES...... 36

Megestrol acetate for treatment of anorexia-cachexia syndrome (Review) i Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. [Intervention Review] Megestrol acetate for treatment of anorexia-cachexia syndrome

Graciela Berenstein2, Zulma Ortiz1

1Epidemiological Research Center, National Academy of Medicine, Buenos Aires, Argentina. 2(Deceased) Epidemiology Department, Hospital Nacional Dr Alejandro Posadas, Buenos Aires, Argentina

Contact address: Zulma Ortiz, Epidemiological Research Center, National Academy of Medicine, Conde 718 10 H, Pacheco de Melo 3081, Buenos Aires, Federal Capital, 1425, Argentina. [email protected]. [email protected].

Editorial group: Cochrane Pain, Palliative and Supportive Care Group. Publication status and date: Edited (no change to conclusions), published in Issue 7, 2011. Review content assessed as up-to-date: 19 August 2007.

Citation: Berenstein G, Ortiz Z. Megestrol acetate for treatment of anorexia-cachexia syndrome. Cochrane Database of Systematic Reviews 2005, Issue 2. Art. No.: CD004310. DOI: 10.1002/14651858.CD004310.pub2.

ABSTRACT

Background

This is an updated version of a previously published review in The Cochrane Library (Issue 2, 2005) on ’Megestrol acetate for the treatment of anorexia-cachexia syndrome’.

Megestrol acetate (MA) is currently used to improve appetite and to increase weight in cancer-associated anorexia. In 1993 MA was approved by the USA’s Federal Drug Administration for the treatment of anorexia, cachexia, or unexplained weight loss in patients with AIDS. The mechanism by which MA increases appetite is unknown, and its effectiveness for anorexia and cachexia in neoplastic and AIDS patients is under investigation.

Objectives

To evaluate the efﬁcacy, effectiveness and safety of MA in palliating anorexia-cachexia syndrome in patients with cancer, AIDS and other underlying pathologies.

Search methods

Studies were sought through an extensive search of the electronic databases, journals, reference lists, contact with investigators and other search strategies outlined in the methods. The most recent search for this update was carried out in June 2006.

Selection criteria

Studies were included in the review if they assessed megestrol acetate compared to placebo or other drug treatments in randomized controlled trials of patients with a clinical diagnosis of anorexia-cachexia related to cancer, AIDS or another underlying pathology.

Data collection and analysis

Data extraction was conducted by two independent review authors, and methodological quality evaluated. Quantitative analyses were performed using appetite and quality of life as a dichotomous variable, and weight gain was analysed as continuous and dichotomous variables. Studies with more than 50% of patients lost to follow-up were excluded from the analysis.

Thirty trials were included in the original review, four new trials were identified for this update, but only two met the inclusion criteria (4123 + 703 patients). Twenty-two trials compared MA at different doses with placebo; five compared different doses of MA versus other drugs; two compared MA with other drugs and placebo; and five compared different doses of MA. For all patient conditions, meta- analysis showed a benefit of MA compared with placebo, particularly with regard to appetite improvement and weight gain in cancer patients. Analysing quality of life, clinical and statistical heterogeneity was found and discussed. There was insufficient information to define the optimal dose of MA.

Authors’ conclusions

This review demonstrates that MA improves appetite and weight gain in patients with cancer. No overall conclusion about Quality of Life (QoL) could be drawn due to heterogeneity. The small number of patients, methodological shortcomings and poor reporting have not allowed us to recommend megestrol acetate in AIDS patients or with other underlying pathologies. Since the last version of this review, none of the new relevant studies have provided additional information on this treatment, in contrast to other therapeutic approaches that should be considered.

PLAIN LANGUAGE SUMMARY

Megestrol acetate for the treatment of anorexia-cachexia syndrome

Megestrol acetate improves appetite and weight gain in patients with anorexia cachexia syndrome related to cancer.Megestrol acetate’s mechanism of action is unknown. There are concerns regarding the possible recommendations for this drug; particularly in the improvement of quality of life in health care and in cancer patients. Quality of life is the cornerstone for delivery of good palliative medicine. The review found that megestrol acetate signiﬁcantly increased appetite and weight gain in cancer patients but there was not enough evidence to reach a conclusion about the effect on quality of life and the optimal dose. There was too little information on AIDS patients or those patients with other underlying pathologies. A low incidence of adverse effects was found.

in 15% to 40% of patients with cancer, and in more than 80% of BACKGROUND patients with advanced illness (Bruera 1992). This review is an update of a previously published review in The Cochrane Library (Issue 2, 2005) on megestrol acetate for anorexia Cachexia in cancer patients is thought to occur as a result of cachexia syndrome. Anorexia cachexia syndrome is a common metabolic changes brought about by substances secreted by the tu- clinical problem that substantially impacts upon the quality of mour and the host (Alexander 1993). Substances have been iden- life and survival of affected patients. It is characterized by loss of tified that cause severe anorexia and weight loss. Tumour necrosis appetite, weight loss and tissue wasting, accompanied by a decrease factor, synthesized by the host’s macrophages (important cells in in muscle mass and adipose tissue, impoverishing the quality of the immune system), and inflammatory cytokines (including in- life, and often preceding the patient’s death (Nelson 1994; Splinter terleukin 1 (I1) and 6 (I6)) are considered responsible for some of 1992). the clinical manifestations (Mantovani 1998). More than two-thirds of patients dying with advanced cancer suf- Early intervention and attention to nutritional status are essential fer from anorexia cachexia syndrome (Argilés 2001). Anorexia in patients with anorexia cachexia syndrome. Pharmacological in- cachexia syndrome is also described in other pathologies such as terventions for neoplastic cachexia include drugs that stimulate the Acquired Immune Deficiency Syndrome (AIDS); anorexia ner- appetite (megestrol acetate and dronabinol); cytokine inhibitors vosa; in degenerative illnesses of the central nervous system; and (such as cyproheptadine, thalidomide, pentoxifylline and an eicos- in terminally ill patients (Von Roenn 1996). Incidence is variable apentaenoic acid (EPA)); and anabolic agents such as nandrolone and difficult to determine but in general the syndrome may occur decanoate, oxandrolone and corticosteroids (Balog 1998). EPA

Megestrol acetate for treatment of anorexia-cachexia syndrome (Review) 2 Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. seems to suppress well-characterized mediators of cancer associ- 1) To evaluate the efficacy, effectiveness and safety of megestrol ated wasting, including interleukin-6, an inflammatory cytokine. acetate in palliating anorexia-cachexia syndrome in sub-groups of It also acts over the proteolysis-inducing factor, another well-de- patients with cancer, AIDS, and other underlying pathologies. scribed mediator (Barber 1999; Wigmore 1997). 2) To determine the optimal dose regimen for megestrol acetate in palliating the anorexia-cachexia syndrome. Megestrol acetate is a synthetic progestogen agent. The biological mechanism of the anti-tumoral activity of megestrol acetate is not well understood but it probably acts on hormone-dependent METHODS tumoral cells. Inhibitory effects on growth in the cellular cycle are not phase-specific, but their activity seems to be maximized in phase G1 of cellular division (Tchekmedyian 1986). Criteria for considering studies for this review Megestrol acetate was first synthesized in England in 1963. De- veloped as an oral contraceptive, the agent was first tested in the treatment of breast cancer in 1967 and later on, for the treat- Types of studies ment of endometrial cancer. Megestrol acetate is currently used Randomized controlled trials (RCTs) which may be double blind, to improve appetite and to increase weight in cancer-associated single blind or unblinded. Cross-over studies were included if they anorexia. From September 1993 megestrol acetate was approved reported the results of the first phase of the study. Both inpatient by the Federal Drug Administration (FDA) in the USA for the and outpatient study settings were included. treatment of anorexia, cachexia, or unexplained weight loss in patients with AIDS. In addition, there are recent reports of the drug being used to improve the quality of life of elderly patients with Types of participants cachexia. A possible role in anorexia nervosa has also been pro- Trials of patients with a clinical diagnosis of anorexia-cachexia re- posed (Yeh 2000). lated to cancer, AIDS or another underlying pathology (independent of gender, age or race) were included. Megestrol acetate is commonly available as a tablet of 80 mg or liquid form (40 mg of micronized megestrol acetate per mL). A great variability in doses is observed in the scientific literature, Types of interventions ranging from 100 mg to 1600 mg per day (Tchekmedyian 1992; The review focuses on the following treatment comparisons: Von Roenn 1994). The liquid form is usually dosed at 20 ml per a) Megestrol acetate at any dose versus placebo; day and the oral one is four tablets per day. The recommended b) Megestrol acetate at any dose versus other active drug treat- duration of treatment is six weeks or more. Megestrol acetate is ments (stimulants of appetite such as dronabinol; cytokine in- considered a relatively non toxic drug with a low incidence of hibitors such as cyproheptadine, EPA and anabolic agents such as adverse effects such as fluid retention, venous thrombosis, diar- nandrolone decanoate and corticosteroids); rhoea, rash, impotence, pruritus, increased blood sugar level, and c) Megestrol acetate at different doses. headache (Loprinzi 1990a; Vadell 1998; Von Roenn 1994).

Types of outcome measures Although the mechanism by which megestrol acetate increases ap- The following outcome measures were assessed: petite is unknown, most hypotheses point to action on cytokines, • appetite increase, expressed as a dichotomous variable which inhibit the action of tumour necrosis factor on fatty tis- (number of patients who experience appetite increase) or a sue and its products. Currently, interest is focused especially on continuous variable (caloriﬁc intake expressed as calories/day); its effectiveness in the treatment of anorexia and cachexia in neo- • weight gain, measured as a dichotomous variable (number plastic and AIDS patients. Studies at the Mayo Clinic and The of patients who gain weight) and as a continuous variable in kg/ North Central Cancer Treatment Group Study have reported and day at the end of the treatment compared with the baseline; reviewed multiple placebo-controlled, randomized, double blind, • measurements of the mid-arm circumference and triceps clinical trials of megestrol acetate and other drugs for the improve- skin fold thickness by anthropometry, as a percentage of the ment of anorexia cachexia syndrome in all types of cancer (Jatoi differences in the total body muscle and fat mass; 2004; Loprinzi 1990a). • improvement in quality of life (QoL), by means of a validated instrument, or with scales of functional scores (e.g., Index of Karnofsky and performance status) that measure the OBJECTIVES well-being status of the patient. The QoL measures will depend

Megestrol acetate for treatment of anorexia-cachexia syndrome (Review) 3 Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. on the instrument used, e.g., patient assessments using a Likert- The results of the search strategy were independently screened type scale based on patients’ statements and self-report by two review authors (EGB, ZO) and assessed for inclusion in questionnaires; or the use of the Spitzer QL-Index of quality of the previous and updated review. Disagreement was resolved by life, completed by the clinician. discussion. Reasons for excluding trials were reported. Study withdrawals and dropouts were analyzed as: • total number of dropouts and withdrawals, Data extraction • number of withdrawals due to lack of effectiveness of Data on patients, methods, interventions, outcomes and results treatment, were extracted by two review authors using a data extraction form • number of withdrawals due to adverse effects. (EGB, ZO) in the previous and updated review. Differences were Adverse effects: these were analyzed as the number of patients who resolved by consensus, and when necessary, in consultation with suffer an event described as a side effect by the authors of each a third review author. study.

Quality of studies Search methods for identification of studies The methodological quality of the studies was evaluated using a validated scale called the Oxford Quality Scale (Jadad 1996). This scale includes an evaluation of the randomisation, blinding and patient attrition. The scale produces a composite score ranging Electronic searches from one (low quality) to five (high quality). The three item scale The following electronic databases were searched to identify rele- is applied as follows: vant studies: • is the study randomised ? If ’yes’ , then one point If • MEDLINE from 1966 to October 2002, subsequent search described, is the randomisation appropriate? If ’yes’ add one June 2006 (see Appendix 1); point, if not deduct one; • • EMBASE from 1986 to 2002, subsequent search, week 28, is the study double blind ? If ’yes’, then one point. Is the 2006 (see Appendix 2); double blind method appropriate ? If ’yes’ add one point, if not deduct one; The Cochrane Central Register of Controlled Trials (CENTRAL), • are withdrawals and dropouts described? (i.e., the number The Cochrane Library Issue 1, 2002, subsequent search Issue 2, and reason for drop-outs for each of the treatment groups). If 2006. ’yes’, add one point. The Cochrane Pain, Palliative and Supportive Care Group Trials Register (June 2006) (see Appendix 3) Allocation concealment was also evaluated as a parameter of quality The general strategy for identifying RCTs in MEDLINE was com- of the design of the studies, and is reported in the ’Characteristics bined with a strategy designed to retrieve trials of megestrol acetate of included studies’ table. for cachexia. For the identification of studies included or considered for this review, detailed search strategies were developed for Data analysis each database searched. Studies with more than 50% of patients lost to follow-up were not included in the analysis. We only analyzed crossover studies that Searching other resources included results from the first treatment period in order to avoid Lists of references of the included studies were checked to identify carry-over effects. further trials. For dichotomous variables, treatment effects were computed as Studies were not excluded on the basis of language or publication relative risk (RR) with 95% confidence intervals (CI). For contin- status (published, unpublished, in press, and in progress). uous variables measured weight gain differences in means and their Additional data from published trials were sought by contacting 95% CI were calculated (weighted mean difference - (WMD) and authors. for quality of life (including different scales), differences in means and their 95% CI were calculated (standardised mean difference - (SMD). Only validated scales with a normal distribution were Data collection and analysis included for the analysis. Validity of the scale was determined by the psychometric properties of the instrument described in the trial by the review authors. A random effect model was used in the analysis. Statistical hetero- Study selection geneity between studies was analysed with a Chi-square test, using

Megestrol acetate for treatment of anorexia-cachexia syndrome (Review) 4 Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. P < 0.1 as a cut-off value to represent the presence of signiﬁcant Excluded studies heterogeneity. When a high level of heterogeneity was detected, In this 2006 update we excluded one additional study (Yeh 2004) attempts were made to identify the sources of the heterogeneity, and one unpublished study (Macbeth 1994) taking the total of and subsequent meta-analysis were performed using a random ef- excluded studies to three. fect model.

Subgroup analysis Included studies Analysis of subgroups was undertaken according to the underlying Further independent assessment by two review authors led to the pathology of the patients. Three subgroups of studies were deﬁned: selection of two new articles: one from Jatoi 2004 and the other • patients with AIDS; from Ulutin 2002. Three abstracts were included (Gambardella • patients with cancer; 1998; Pardo 2003; Zeca 1995), but two had only a small amount • patients with other underlying disease. of data, Gambardella 1998 had no data at all although it met the inclusion criteria, however, as there was no data available at present it was excluded from the analyses for now and should data become available at a later date, it will be included in a future update. Sensitivity analysis

In order to explore the impact of specific factors on the meta- Many of these citations were replicated across the three databases, analysis results, sensitivity analyses was undertaken with: and five studies that appeared to meet the inclusion criteria were • studies of high methodological quality (defined as studies subsequently found to be follow-up reports or duplicate publica- with appropriate concealment of allocation, appropriate tions (see Characteristics of excluded studies’ table). blinding, and analysis by intention-to-treat (ITT); A total of 34 reports (describing 35 trials) fully met the inclusion • studies where patients received more than six weeks of criteria for this update and provided data for analysis. treatment. The designs of the 34 trials were as follows:

The statistical analyses were carried out using the statistical pack- age, RevMan Analyses 1.0.2, in RevMan 4.2.10. Megestrol acetate at different doses compared with placebo Twenty-two trials compared megestrol acetate at different doses with placebo (Beller 1997; Bruera 1990; Bruera 1998; De Conno 1998; Erkurt 2000; Eubanks 2002; Feliu 1992; Fietkau 1996; RESULTS Loprinzi 1990b; Marchand 2000; McMillan 1994; McQuellon 2002; Oster 1994; Rowland 1996; Schmoll 1992; Tchekmedyian 1992; Vadell 1998; Von Roenn 1994; Weisberg 2002; Westman Description of studies 1999; Yeh 2000; Zeca 1995). See: Characteristics of included studies; Characteristics of excluded studies. Searching electronic databases identified: Megestrol acetate at different doses compared with other • 99 reports in MEDLINE (from 1966 to October 2002, the treatment drugs subsequent search identified 24 reports in MEDLINE (from Five trials compared different doses of megestrol acetate with other 2002 to July 2006); drug treatments. Megestrol acetate was compared with dronabi- • 164 reports in EMBASE (from 1986 to 2002), the nol in two studies (Jatoi 2002; Timpone 1997); dexamethasone subsequent search identified 24 in EMBASE (from 2002 to week and fluoximesterone in one study (Loprinzi 1999); nandrolone 28, 2006); and decanoate in one study (Batterham 2001) and eicosapentaenoic • 71 in the Cochrane Central Register of Controlled Trials acid (EPA) in one study (Jatoi 2004). (CENTRAL), (Issue 2, 2006), the subsequent search identified 59 reports (in Issue 1, 2002). Megestrol acetate at different doses compared with other Three abstracts retrieved from handsearching in CCRT (CEN- treatment drugs and placebo TRAL) were identified in this updated review. We attempted to contact the authors of these abstracts for further trials data but Two studies compared megestrol acetate with other drugs and with no success. placebo (Lai 1994 (prednisolone); Chen 1997 (cisapride)).

Megestrol acetate for treatment of anorexia-cachexia syndrome (Review) 5 Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Megestrol acetate at different doses Dose Five trials (compared different doses of megestrol acetate (Gebbia Across the studies, the dose of megestrol acetate ranged from 100 1996; Heckmayr 1992; Loprinzi 1994; Pardo 2003; Ulutin 2002). mg per day to 1600 mg per day in at least one of the study arms. The included studies were categorized according to the health care The doses of megestrol acetate assessed were as follows: problem of the patient - see Table 1 for a summary. • 400 mg per day or less Seventeen trials: Batterham 2001(400 mg per day); Beller 1997 (160 mg per day); Chen 1997 (160 mg per day); De Conno 1998 Patient characteristics (320 mg per day); Feliu 1992 (240 mg per day); Fietkau 1996 (160 mg per day); Gebbia 1996 (160 mg and 320 mg per day); A total of 4826 patients recruited were included in this update. Heckmayr 1992 (160 mg per day); Lai 1994 (160 mg per day); The mean age of patients included in the treatment and control Loprinzi 1994 (160 mg per day); Pardo 2003 (320 mg per day); groups across all studies was 56 years. The proportion of males Timpone 1997 (250 mg per day); Ulutin 2002 (160 mg and 320 to females in the treatment groups was 1343/562, compared to mg per day); Vadell 1998 (160 mg per day); Von Roenn 1994 2038/833 in the control groups. (100 mg and 400 mg per day); Westman 1999 (320 mg per day); Zeca 1995 (320 mg per day). • 480 mg per day Patients with any cancer Nine trials: Beller 1997; Bruera 1990; Bruera 1998; Erkurt 2000; Twenty-six trials (total number of patients 4148) (Beller 1997; Heckmayr 1992; Loprinzi 1994; McMillan 1994; Schmoll 1992; Bruera 1990; Bruera 1998; Chen 1997; De Conno 1998; Erkurt Vadell 1998 2000; Feliu 1992; Fietkau 1996; Gebbia 1996; Heckmayr 1992; • 600 mg per day Jatoi 2002; Jatoi 2004; Lai 1994; Loprinzi 1990b; Loprinzi 1994; Loprinzi 1999; McQuellon 2002; McMillan 1994;Pardo 2003; Two trials: Jatoi 2004; Pardo 2003 Rowland 1996; Schmoll 1992; Tchekmedyian 1992; Vadell 1998; • 750 to 800 mg per day Westman 1999; Ulutin 2002; Zeca 1995) assessed the effective- Eleven trials: Jatoi2002; Timpone 1997; Loprinzi 1990b; Loprinzi ness of megestrol acetate for anorexia-cachexia syndrome in cancer 1994; Loprinzi 1999; McQuellon 2002; Oster 1994; Rowland patients where the primary site was: 1996; Von Roenn 1994; Weisberg 2002; Yeh 2000 • lung cancer (1792 patients, 43%); • 1280 mg per day • gastrointestinal and pancreatic cancer (1045 patients, 25%); • head and neck cancer (347 patients, 8%); One trial: Loprinzi 1994 • gynecological cancer (74 patients, 2%); • 1600 mg per day • non-specified sites (890 patients, 21%). One trial: Tchekmedyian 1992 • One trial in children with cystic fibrosis assessed megestrol acetate at a dose of 10 mg/kg per day (Marchand 2000) Patients with AIDS Four trials (total number of patients, 435) assessed the effectiveness of megestrol acetate for anorexia-cachexia syndrome in AIDS pa- Study duration tients (Batterham 2001; Oster 1994; Timpone 1997; Von Roenn The study duration ranged from ten days to 24 weeks. One study 1994) ran for two years (Rowland 1996) and measured survival and response to chemotherapy as well as the effect of megestrol acetate on anorexia-cachexia syndrome. The median trial duration time Patients with other underlying conditions was eight weeks. Five trials had a duration of more than 12 weeks, (see ’Characteristics of included studies’ table). Four trials (total number of patients, 243) assessed the effectiveness • Assessment at seven days (Bruera 1990) of megestrol acetate for anorexia-cachexia syndrome in patients • Assessment at ten days (Bruera 1998) with the following conditions: • Final assessment at two weeks (Beller 1997; De Conno • chronic obstructive pulmonary disease (COPD): one trial 1998; Zeca 1995) (Weisberg 2002) of 145 patients; • Assessment at three weeks (Lai 1994) • cystic fibrosis: two trials (Marchand 2000 of 12 patients • Assessment at four weeks/one month (Chen 1997; Gebbia and Eubanks 2002 of 17 patients); 1996; Heckmayr 1992; Jatoi 2002; Loprinzi 1990b; Loprinzi • elderly: one trial (Yeh 2000) of 69 patients. 1999; Pardo 2003)

Megestrol acetate for treatment of anorexia-cachexia syndrome (Review) 6 Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. • Assessment at six weeks (Fietkau 1996; Tchekmedyian To address the problem of patient attrition and loss to follow- 1992) up, both ITT and per protocol (PP) analyses were undertaken for • Assessment at eight weeks/two months (Chen 1997; Feliu each of the outcomes. Three trials were excluded from the analysis 1992; Loprinzi 1994; Schmoll 1992; Weisberg 2002) (Jatoi 2002; Jatoi 2004; Westman 1999) because more than 50% • Assessment at 12 weeks/three months (Batterham 2001; of patients were lost to follow-up. In addition, the cross-over trials Erkurt 2000; Fietkau 1996; Jatoi 2004; Marchand 2000; conducted by Bruera et al (Bruera 1990; Bruera 1998) did not McMillan 1994; McQuellon 2002; Oster 1994; Timpone 1997; report the results from the first treatment period and, because Ulutin 2002; Vadell 1998; Von Roenn 1994; Westman 1999; of the problem of carry-over effects, the data from these studies Yeh 2000) were not included. In some cases it was not possible to perform • Assessment at six months or more (Eubanks 2002; Rowland meta-analysis due to lack of information (Batterham 2001; Beller 1996) 1997; Chen 1997; De Conno 1998; Gebbia 1996; Oster 1994; Tchekmedyian 1992; Weisberg 2002; Westman 1999; Yeh 2000; ). Risk of bias in included studies Only three of five intended outcome measures were analyzed: The methodological quality of the included studies was assessed 1) appetite increase, expressed as a dichotomous variable (number using the Oxford Quality Scale (Jadad 1996). Each report was of patients who experience appetite increase), scored independently for quality by the review authors using the 2) Weight gain, expressed both as a dichotomous and continuous three-item scale described in the ’Methods’ section above who variable in kg/day at the end of the treatment compared with the agreed a ’consensus’ score. The scores for methodological quality baseline, and were generally high. 3) Improvement in quality of life (QoL) analysed as a dichoto- Twenty one trials (62%) scored three or more out of a maximum mous variable (number of patients who experienced improved of five: QoL, measured in different and validated scales). Basically, qual- Beller 1997; Bruera 1990; Bruera 1998; De Conno 1998; ity of life was assessed by different scales and derived by different Eubanks 2002; Feliu 1992; Fietkau 1996; Jatoi 2002; Jatoi 2004; methods. Loprinzi 1990b; Loprinzi 1994; Loprinzi 1999; Marchand 2000; No data were available for the analysis of mid-arm circumference, McQuellon 2002; Oster 1994; Tchekmedyian 1992; Vadell 1998; triceps skin fold thickness, appetite increase (calorific intake ex- Von Roenn 1994; Weisberg 2002; Westman 1999; Yeh 2000. pressed as calories/day) and QoL expressed as a continuous vari- Thirteen trials (38%) achieved a low score (two points or able. lower): Batterham 2001; Chen 1997; Erkurt 2000; Gebbia 1996; Heckmayr 1992; Lai 1994; McMillan 1994; Pardo 2003; Rowland 1996; Schmoll 1992; Timpone 1997; Ulutin 2002; Zeca 1995. Megestrol acetate versus placebo The overall results have shown improvement in appetite for pa- Effects of interventions tients treated with megestrol acetate (RR = 2.76 (95% CI 1.65 to 4.61)) (Comparison 01 01). The only subcategory that could be Data from the included studies were meta-analysed in to three analyzed was cancer patients (Loprinzi 1990b; Erkurt 2000; Feliu groups: 1992; Lai 1994; Schmoll 1992; Zeca 1995). Both subcategories • megestrol acetate versus placebo, AIDS and other underlying pathologies could not be analyzed be- • megestrol acetate versus other active drug treatments, cause there was only one trial included in each one. • megestrol acetate at different doses. For cancer patients a statistically significant benefit was described in appetite improvement (RR 3.03, 95% CI 1.83 to 5.01) and they were further categorized into: weight gain as a continuous variable (WMD = 3.56 (95% CI 1.27 • patients with cancer, to 5.85)) (Comparison 01 02). The same direction and significance • patients with AIDS, of the results was seen with weight gain as a dichotomous variable • patients with other underlying pathologies. (RR = 2.14 (95% CI 1.41 to 3.24) (Comparison 01 03). The two new complete studies and two abstracts could not be Six trials (four on cancer sub-category and two on AIDS) were included in the analysis because of the drop-outs in Jatoi 2004, included for the comparison of the effect of megestrol acetate and the comparison of two low doses of megestrol in Ulutin 2002, placebo on health-related quality of life (HRQoL) and no signif- the comparison of one low dose of megestrol versus high dose in icant results were seen. Clinical and statistical heterogeneity were Pardo 2003 as explained below, and the lack of information in detected and explained by severity of illness, treatment duration Gambardella 1998. Only Zeca 1995 was included in the analysis and different scales (P < 0.0001) (Comparison 01 04). Three tri- of appetite. als in the subcategory of patients with cancer have used the per-

Megestrol acetate for treatment of anorexia-cachexia syndrome (Review) 7 Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. formance status and Karnofsky Index (KI), and one the Linear In order to perform the meta-analysis, we defined 400 to 480 mg/ Analog Self Assessment, nine items (LAS). Studies included in the d as the cut off value. This was the most frequently reported dose AIDS subcategory used KI and LAS. by authors. Then, we compared this dose to high doses (=>800 A random effect analysis was carried out and a P value < 0.1 was mg/d) and to low doses (<400 to 480 mg/d). In the first group only found. A single outlier study (Erkurt 2000) can make a very large two trials could be analysed (Loprinzi 1994; Schmoll 1992). In the contribution to the statistical heterogeneity, resulting in a conclu- second group two trials, out of four, were included in the analysis sion that there is heterogeneity in both appetite and weight gain (Heckmayr 1992; Loprinzi 1994). We excluded the Gebbia 1996 when the studies yield a “credible” conclusion. The pooled RR may and Ulutin 2002 studies from this analysis because they used only be affected by Erkurt’s study, which had an enormous treatment low doses and Pardo 2003 did not meet the criteria to include in effect. The pooled value in fact is dissimilar to individual values the analysis. The results of the meta-analysis were not significant. from all studies. Erkurt´ s quality study was low. The analysis and interpretation of data were difficult to understand and regarding Study withdrawals and dropouts QoL was excluded from the analysis due to the studies using an un-validated instrument for its measurement. Only the studies who reported the total number of drop-outs and Loprinzi 1990b, showed the largest pooled RR when we looked at withdrawals in each arm were included. There were 224/1115 in the analysis of weight gain as a dichotomous variable. The authors the megestrol acetate group and 207/886 in the placebo group of reported that a percentage of patients gained at least 15 lb after all the analyzed categories. Only one study (Oster 1994) reported they were given a high dose of MA. The data were obtained by two withdrawals due to lack of efficacy of treatment. A total of 29/ sources: patient-recorded home weights and institution weights. 1596 in the treated group and 24/1143 in the placebo group were We have included the highest level of weight gain reported . This withdrawals due to adverse effects in this update including the study was given five points as a quality score. study of Ulutin 2002. The number of patients who suffer any side effect was analyzed and those studies that reported toxicity scores were also checked. The two abstracts included reported no side- Megestrol acetate versus other drugs effects. Five of seven identified studies comparing megestrol acetate with Drug safety, impotence in men, edema of the lower limbs, deep other drugs were pooled for the analysis. Three trials in cancer vein thrombosis and gastrointestinal intolerance were the most subcategory (Chen 1997; Lai 1994; Loprinzi 1999) and two in frequently reported adverse effects observed in the studies. None AIDS (Batterham 2001; Timpone 1997). Loprinzi et al have com- of the differences between treatment and placebo groups were pared megestrol acetate to fluoxymesterone and dexamethasone. found to be statistically significant, except for the occurrence of To analyse this trial we divided the total number of patients in- edema, which occurred with greater frequency in patients receiving cluded in the megestrol group by two. In other words, the num- megestrol acetate (RR = 1.74 (95% CI 1.29 to 2.35) (Comparison ber of megestrol patients in each comparison was taken to be 79 01 05). instead of 158. This method allowed us to perform the analysis. Jatoi 2002 and Jatoi 2004 were excluded due to the high rate of drop-outs. Sensitivity analysis When we looked at the overall results, megestrol acetate did not This 2006 update does not show any change to the sensitivity show benefits in terms of appetite improvement in comparison analysis from the previous review. to other drugs (RR = 1.15 (95% CI 0.75 to 1.76) (Comparison The sensitivity analysis was undertaken with studies where cancer 02 01), and results were inconclusive for weight gain (RR = 1.29 patients received more than six weeks of megestrol acetate versus (95% CI 0.94 to 1.96) (Comparison 02 03). placebo. Two outcomes were analyzed: appetite improvement and Two studies (Lai 1994; Loprinzi 1999a) included in the analysis weight gain. We observed an overall benefit in both outcomes of have measured HRQOL as an outcome using two different in- the treated group. Appetite improvement showed a RR = 3.21 struments. The first one used the K I and Loprinzi used Spitzer (95% CI 1.54 to 6.70) (Comparison 03 01) and weight gain a RR QL index. No significant results were found (RR = 1.08 (95% CI = 1.86 (95% CI 1.31 to 2.63) (Comparison 03 02). 0.80 to 1.45) (Comparison 02 04). Another subgroup analysis was performed considering methodological quality of the studies. The group with high Oxford Qual- There was not enough data to compare megestrol acetate versus ity Scores (Quality Score of three, four or five) showed a benefit other drugs in HIV/AIDS and other underlying pathology pa- for the megestrol acetate in the outcome weight gain only, both as tients. a dichotomous and continuous variable (RR = 1.66 (95% CI 1.13 to 2.44) (Comparison 03 06); WMD = 1.87 (95% CI 1.20 to 2.54) (Comparison 03 05). For the outcome of appetite improve- Different dose levels of megestrol acetate ment there were not enough studies with good Oxford Quality

Megestrol acetate for treatment of anorexia-cachexia syndrome (Review) 8 Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Scores. In the group of low methodological score, we observed an methodological issues related to QoL measurement relevant to ad- overall benefit in the treatment group for the outcome appetite dress in order to explain clinical heterogeneity: improvement (RR = 3.63 (95% CI 2.06 to 6.39) (Comparison 03 a) different types of instruments, 04) and weight gain (RR = 2.49 (95% CI 1.45 to 4.27) (Compar- ison 03 06). b) extensive testing to assess global QoL, c) difficulty administering this testing led to inconclusive QoL findings. DISCUSSION On the other hand some patients cannot complete quality of life measures because they have cognitive impairments, communica- This updated review does not provide any additional information tion deficits, are in severe distress, or because the measures are too on this treatment as the two new studies that were included did burdensome. The studies included in this update used ten types not provide data of any significance Jatoi 2004; Ulutin 2002). of instruments with different psychometric properties. Two systematic reviews have been undertaken recently. A systematic review by Maltoni 2001 assessed the efficacy of high dose progestins for the treatment of anorexia cachexia syndrome in pa- Another factor which may explain the discordance between ap- tients with hormone-independent tumours. This review included petite improvement and the lack of benefit on QoL, centers on 15 randomized clinical trials and more than 2000 patients. The the fact that multiple symptoms converge among advanced can- authors concluded that high-dose progestins improve appetite and cer patients. Donnelly et al (Donnelly 1995) evaluated symptom weight, but could not define optimal dose, duration of treatment prevalence in 1000 such patients and found that although anorexia or the impact on quality of life. The review by Ruiz-García et had a high prevalence, it was not an isolated symptom. Nausea, al. (Ruiz-Garcia 2002) evaluated the efficacy of megestrol acetate fatigue, dry mouth, pain, and a variety of other symptoms accom- versus placebo in patients with cancer and anorexia cachexia syn- pany anorexia. drome. Eight trials (719 patients) were included in the review and Sensitivity analysis showed that the poorest quality trials over- found that megestrol acetate was associated with a slight weight estimated positive results. gain at doses of 240 mg per day or less. No statistically significant effect was observed with higher doses.

The aim of the present update was to assess the efficacy, effec- Finally, a more systematic approach to the measurement of tiveness and safety of megestrol acetate for the management of HRQOL in these patients would be helpful in better understand- anorexia-cachexia syndrome, a common clinical problem that sub- ing and analyzing treatment effects and the impact of megestrol stantially impacts upon the quality of life and survival of affected acetate on patient QoL. patients. Although the number of randomised trials retrieved was 71, only 34 of them were included for the analysis and 12 received Although we included studies which had up to 50% dropouts, we a high quality score (quality scale score of four or five). The to- found a particularly high dropout rate probably due to advanced tal number of patients was 4826. However, we could not include cancer. The population studied is very sick, so they tend to suffer more than 50% of the studies in the analysis. Data was available substantial cancer anorexia/cachexia along with many co morbid from only 19 studies. problems linked to advanced cancer. The adverse event profile of megestrol acetate has shown edema to be the only one significant difference between placebo and mege- There was consistency between this review and results reported by strol acetate, suggesting that megestrol acetate is a safe treatment the majority of other similar studies. This review confirms that option in these patients. Megestrol acetate is a good choice to increase appetite and weight gain in cancer patients compared to placebo. The improvement of the nutritional status by anthropometric changes could not be AUTHORS’ CONCLUSIONS assessed due to lack of information. Although some authors have suggested that there may be a ten- Implications for practice dency towards greater efficacy with higher doses, this update New trials located for this update have not provided additional showed that clinical effects of megestrol acetate appear not to be information to the original conclusion. Megestrol acetate may be dose-related. prescribed in patients with cancer to increase appetite and weight When we analysed the improvement of QoL we found great gain. At the moment, there is no evidence to recommend megestrol heterogeneity between studies. Jatoi 2000 had described many acetate to improve quality of life (QoL). This update has followed

Megestrol acetate for treatment of anorexia-cachexia syndrome (Review) 9 Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. the Cochrane Collaboration Guidelines to perform an unbiased Federal Drug Administration has approved megestrol acetate to review. However, the results of a meta-analysis are influenced by be used in AIDS patients more research is needed. the quality of the primary studies included. Quality is difficult to define. It could address the design, conduct and analysis of a trial, its clinical relevance, or the quality of reporting. Studies of low methodological quality can alter the interpretation of the benefit ACKNOWLEDGEMENTS of intervention. In this update 68% of the studies were assessed as moderate or low quality trials. We would like to thank the Iberoamerican Cochrane Collabora- tion Centre for their support, particularly Marta Roque for her Many concerns are still to be resolved. HRQoL is an important critical review and Gerard Urrutia for his comments. We are grate- goal in health care and cancer clinical trials and is the cornerstone ful to Dr CL Loprinzi and Paul Novotny from the Mayo Clinic; for delivery of good palliative medicine. The increasing recogni- Professor E Bruera, Department of Palliative Care and Rehabili- tion of patient autonomy means that subjective measures will be- tation Medicine at Cross Cancer Institute, University of Alberta, come more important and, in the current climate of evidence- Edmonton, and MD Anderson for supplying additional data for based medicine, such measures must be valid and reliable. this review. We would also like to thank Marcelo García Diéguez and Nicolas Garrigue from Argentina who have helped us with the Implications for research search strategy and comments. We would like to thank Sylvia Bick- This update has shown that there is still a need for high quality ley (Trials Search Co-ordinator, Cochrane Pain, Palliative Care & studies on the effectiveness of megestrol acetate in some of the Supportive Care Group) for performing the bibliographic searches outcomes we have reviewed, e.g. HRQOL. Even though the USA’s in the present version of this update.

REFERENCES

References to studies included in this review controlled clinical trial. European Journal of Cancer 1998; 34:1705–9. Batterham 2001 {published data only} Erkurt 2000 {published data only} Batterham MJ, Garsia R. A comparison of megestrol acetate, Erkurt E, Erkisi M. Supportive treatment in weight- nandrolone decanoate and dietary counselling for HIV losing cancer patients due to the additive adverse effects associated weight loss. International Journal of Andology of radiation treatment and/or chemotherapy. Journal of 2001;24(4):232–40. Experimental Clinical Cancer Research 2000;19(4):431–9. Beller 1997 {published data only} Eubanks 2002 {published data only} Beller E, Tattersall M. Improved quality of life with Eubanks V, Koppersmith N. Effects of megestrol acetate on megestrol acetate in patients with endocrine-insensitive weight gain, body composition, and pulmonary function in advanced cancer: A randomised placebo-controlled trial. patients with cystic ﬁbrosis. Journal of Pediatrics 2002;140 Annals of Oncology 1997;8:277–83. (4):439–44. Bruera 1990 {published data only} Feliu 1992 {published data only} Bruera E, Mc Millan K. A controlled trial of megestrol Feliu J, Gonzalez-Baron M. Usefulness of megestrol acetate acetate on appetite, caloric intake, nutritional status, and in cancer cachexia and anorexia. A placebo controlled study. other symptoms in patients with advanced cancer. Cancer American Journal of Clinical Oncology 1992;15(5):436–40. 1990;66:1279–82. Fietkau 1996 {published data only} Bruera 1998 {published data only} ∗ Fietkau R, Riepl M. Supportive treatment with megestrol Bruera E, Scott E. Effectiveness of megestrol acetate in acetate during radio (chemo) therapy. A randomised trial. patients with advanced cancer: a randomized, double-blind, Strahlenther Onkol 1996;172:162–8. crossover study. Cancer Prevention & Control 1998;2(2): Fietkau R, Riepl M, Kettner H, Hinke A, Sauer R. 74–8. Supportive use of megestrol acetate in patients with head Chen 1997 {published data only} and neck cancer during radio(chemo)therapy. European Chen Hui-Chun, Wan Leung S. Effect of megestrol acetate journal of cancer (Oxford, England: 1990) 1997;33(1):75–9. and prepulsid on nutritional improvement in patients [: CN–00137738] with head and neck cancers undergoing radiotherapy. Gambardella 1998 {published data only} Radiotherapy and Oncology 1997;43:75–9. Gambardella A, Pesce L, Bolognino P,Lombardi G, Barbieri De Conno 1998 {published data only} M, Rinaldi C. Megestrol acetate prevents cachexia in elderly De Conno F, Martini C. Megestrol acetate for anorexia cancer patient. Annals of oncology 1998;9, Suppl 3:72. [: in patients with far-advanced cancer: a double-blind CN–00305911]

Megestrol acetate for treatment of anorexia-cachexia syndrome (Review) 10 Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Gebbia 1996 {published data only} McMillan 1994 {published data only} Gebbia V, Testa A. Prospective randomised trial of two dose Mc Millan DC, Simpson JM. Effect of megestrol acetate levels of megestrol acetate in the management of anorexia- on weight loss, body composition and blood screen of cachexia syndrome in patients with metastatic cancer. gastrointestinal cancer patients. Clinical Nutrition 1994;13: British Journal of Cancer 1996;73:1576–80. 85–9. Heckmayr 1992 {published data only} McQuellon 2002 {published data only} Heckmayr M, Gatzemeier U. Treatment of cancer weight Mc Quellon RP, Moose DB. Supportive use of megestrol loss in patients with advanced lung cancer. Oncology 1992; acetate (MEGACE) with head/neck and lung cancer 49(suppl 2):32–4. patients receiving radiation therapy. International Journal of Radiation Oncology Biology and Physics 2002;52:1180–5. Jatoi 2002 {published data only} Oster 1994 {published data only} Jatoi A, Windschitl HE. Dronabinol versus megestrol Oster MH, Enders SR. Megestrol acetate in patients with acetate versus combination for cancer-associated anorexia: AIDS and cachexia. Annals of Internal Medicine 1994;121: A North Central Cancer Tratment Group Study. Journal of 400–8. Clinical Oncology 2002;20(2):567–73. Pardo 2003 {published data only} Jatoi 2004 {published data only} ∗ Pardo J, Mena AM, Montsech L. Megestrol acetate for Jatoi A, Rowland K, Loprinzi CL, et al.An eicosapentaenoic anorexia in lung cancer patients undergoing radiation acid supplement versus megestrol acetate versus both for therapyA randomized trial comparing the efficacy of two patients with cancer-associated wasting: a north central different doses in 130 patients. Proceedings for the American cancer treatment group and National Cancer Institute of Society for Clinical Oncology 2003;22 (abstr 3076):765. [: Canada Collaborative Effort. Journal of Clinical Oncology CN–00485363] 2004;22:2469–76. Rowland 1996 {published data only} Lai 1994 {published data only} Rowland KM, Loprinzi CL. Randomized double-blind Lai YL, Fang FM. Management of anorexic patients in placebo-controlled trial of cisplatin and etoposide plus radiotherapy: A prospective randomized comparison of megestrol acetate/ placebo in extensive-stage small-cell lung megestrol and prednisolone. Journal of Pain and Symptom cancer: A North Central Cancer Treatment Group Study. Management 1994;9:265–8. Journal of Clinical Oncology 1996;14:135–41. Loprinzi 1990b {published data only} Schmoll 1992 {published data only} ∗ ∗ Loprinzi CL, Ellison NM. Controlled trial of megestrol Schmoll E. Risks and benefits of various therapies for acetate for the treatment of cancer anorexia and cachexia. cancer anorexia. Oncology 1992;49:43–5. Journal of the National Cancer Institute 1990;82:1127–32. Schmoll E, Wilke H, Thole R, et al.Megestrol acetate in cancer cachexia. Seminars in oncology 1991;18(1 Suppl 2): Loprinzi 1994 {published data only} 32–4. [: CN–00072963] ∗ Loprinzi CL, Bernath AM. Phase III evaluation of 4 doses Tchekmedyian 1992 {published data only} of megestrol acetate for patients with cancer anorexia and/ Tchekmedyian NS, Hickman M. Megestrol acetate in or cachexia. Oncology 1994;51:2–7. cancer anorexia and weight loss. Cancer 1992;69:1268–74. Loprinzi CL, Michalak Jc, Schaid DJ, et al.Phase III evaluation of four doses of megestrol acetate as therapy for Timpone 1997 {published data only} patients with cancer anorexia and/or cachexia. Journal of Timpone JG, Wright DJ. The safety and pharmacokinetics clinical oncology: official journal of the American Society of of single-agent and combination therapy with megestrol Clinical Oncology 1993;11(4):762–7. [: CN–00092689] acetate and dronabinol for the treatment of HIV wasting syndrome. AIDS Research and Human Retroviruses 1997;13: Loprinzi 1999a {published data only} 305–15. ∗ Loprinzi CL, Kugler JW. Randomized comparison Ulutin 2002 {published data only} of megestrol acetate versus dexamethasone versus Ulutin HC, Arpaci F, Pak Y. Megestrol acetate for cachexia fluoxymesterone for the treatment of cancer anorexia/ and anorexia in advanced non-small cell lung cancer: a cachexia. Journal of Clinical Oncology 1999;17:3299–306. randomized study comparing two different doses. Tumori Loprinzi 1999b {published data only} 2002;88:277–80. Loprinzi CL, Kugler JW. Randomized comparison Vadell 1998 {published data only} of megestrol acetate versus dexamethasone versus Vadell C, Segui MA. Anticachectic efficacy of megestrol fluoxymesterone for the treatment of cancer anorexia/ acetate at different doses and versus placebo in patients with cachexia. Journal of Clinical Oncology 1999;17:3299–306. neoplastic cachexia. American Journal of Clinical Oncology Marchand 2000 {published data only} 1998;21:347–51. Marchand V, Baker SS. Randomized, double-blind, Von Roenn 1994 {published data only} placebo-controlled pilot trial of megestrol acetate in Von Roenn JH, Armstrong D. Megestrol acetate in patients malnourished children with cystic fibrosis. Journal of with AIDS-related cachexia. Annals of Internal Medicine Pediatric Gastroenterology and Nutrition 2000;31:264–9. 1994;121:393–9.

Megestrol acetate for treatment of anorexia-cachexia syndrome (Review) 11 Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Weisberg 2002 {published data only} Balog 1998 Weisberg J, Wanger J. Megestrol acetate stimulates weight Balog DL, Epstein ME, Amidio-Groton MI. HIV wasting gain and ventilation in underweight COPD patients. Chest syndrome: Treatment update. Annals of Pharmacotherapy 2002;121(4):1070–8. 1998;32(4):446–58. Westman 1999 {published data only} Barber 1999 Westman G, Bergman B. Megestrol acetate in advanced, Barber MD, Ross JA, Voss AC, et al.The effect of an oral progressive, hormone-insensitive cancer. Effects on nutritional supplement enriched with fish oil on weight-loss the quality of life: a placebo-controlled, randomised, in patients with pancreatic cancer. British Journal of Cancer multicentre trial. European Journal of Cancer 1999;35: 1999;81:80–6. 586–95. Bruera 1992 Yeh 2000 {published data only} Bruera E. Clinical management of anorexia and cachexia in Yeh SS, Wu SY. Improvement in quality of life measures and patients with advanced cancer. Oncology 1992;49(2):35–42. stimulation of weight gain after treatment with megestrol Donnelly 1995 acetate oral suspension in geriatric cachexia: Results of Donnelly S, Walsh D. The symptoms of advanced a double-blind, placebo-controlled study. Journal of the cancer: Identification of clinical and research priorities by American Geriatric Society 2000;48(5):485–92. assessment of prevalence and severity. Journal of Palliative Zeca 1995 {published data only} Care 1995;22:27–32. Zeca E, Martini C, Venturino P, Tedeschi M, Ventafrida Jadad 1996 V, De Conno F. Efficacy of Megestrol Acetate on Jadad A, Moore RA, Carrol D, et al.Assessing the quality of Anorexia in patients with advanced non hormone-related reports on randomized clinical trials: Is blinding necessary?. tumors: A double-blind placebo controlled clinical trial. Controlled Clinical Trials 1996;170:1–12. European Journal of Cancer 1995;31 A:S261–S262. [: Jatoi 2000 CN–00519065] Jatoi A, Kumar S. On appetite and its loss. Journal of References to studies excluded from this review Clinical Oncology 2000;18:2930–2. Loprinzi 1990a Macbeth1994 {published data only (unpublished sought but not used)} Loprinzi C, Ellison NM, Goldberg RM, et al.Alleviation of Macbeth FR, Gregor A, Cottier B. A randomised study of cancer anorexia and cachexia: Studies of the Mayo Clinic megestrol acetate (MA) and prednisolone (P) for anorexia and the North Central Cancer Treatment Group. Seminars and weight loss in patients with lung cancer. Cochrane in oncology 1990;17(6 suppl 9):8–12. Database of Systematic Reviews 2005, issue 2. [DOI: Loprinzi 1999 10.1002/14651858.CD004310.pub2] Loprinzi CL, Kugler JW, Sloan JA. Randomized comparison McMillan 1999 {published data only} of Megestrol Acetate versus dexamethasone versus Mc Millan DC, Wigmore SJ. A prospective randomised fluoxymesterone for the treatment of cancer anorexia/ study of megestrol acetate and ibuprofen in gastrointestinal cachexia. Journal of Clinical Oncology 1999;17:3299–306. cancer patients with weight loss. British Journal of Cancer Maltoni 2001 1999;79:495–500. Maltoni M, Nanni O, Scarpi E, et al.High-dose progestins Yeh 2004 {published data only} for the treatment of cancer anorexia-cachexia syndrome: A ∗ Yeh S, Hafner A, Chang C, et al.Risk Factors Relating systematic review of randomised clinical trials. Annals of Blood Markers of Inflammation and Nutritional Status to Oncology 2001;12(3):289–300. Survival in Cachectic Geriatric Patients in a Randomized Mantovani 1998 Clinical Trial. Journal of the American Geriatrics Society Mantovani G, Maccio A, Lai P, et al.Cytokyne activity in 2004;52:1708–12. cancer-related anorexia/cachexia: role of megestrol acetate Additional references and medroxyprogesterone acetate. Seminars in oncology 1998 (US);25(2 suppl 16):45–52. Alexander 1993 Nelson 1994 Alexander HR, Norton JA. Pathophisiology of cancer Nelson KA, Walsh D, Sheehan FA, et al.The cancer cachexia. In: Doyle D, Hanks GWC, MacDonald N editor anorexia-cachexia syndrome. Journal of Clinical Oncology (s). Textbook of palliative medicine. 3rd Edition. Oxford: 1994;12:213–25. Oxford University Press, 1998. Ruiz-Garcia 2002 Argilés 2001 Ruiz-Garcia V, Juan O, Perez Hoyos S, et al.Megestrol Argilés JM, Meijsing SH, Pallares-Trujillo J. Cancer acetate: a systematic review usefulness about the weight cachexia. A therapeutic approach. Medical Research Review gain in neoplastic patients with cancer. Medicina Clinica (US) 2001;21:83–101. (Barcelona) 2002;119(5):166–70.

Megestrol acetate for treatment of anorexia-cachexia syndrome (Review) 12 Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Splinter 1992 with HIV: lessons for the oncologist. Oncology 1996;10(7): Splinter TA. Cachexia and cancer: a clinician´ s view. 1049–56. Annals of Oncology 1992;3(3):25–7. Tchekmedyian 1986 Wigmore 1997 Tchekmedyian NS, Tait N, Aisner J. High dose megestrol Wigmore SJ, Fearon KCH, Maingay JP, et al.Down- acetate in the treatment of postmenopausal women with regulation of the acute-phase response in patients with advanced breast cancer. Seminars in oncology 1986;13(4): pancreatic cancer cachexia receiving oral eicosapentaenoic 20–5. acid is mediated via suppression of interleukin-6. Clinical Von Roenn 1996 Science 1997;92:215–21. Von Roenn JH, Knopf K. Anorexia/cachexia in patients ∗ Indicates the major publication for the study

Characteristics of included studies [ordered by study ID]

Batterham 2001

Methods Randomised controlled In a tertiary referral hospital, Sydney

Participants 15 HIV pts a) 4 M mean age 46 yrs b) 6 M mean age 44 yrs c) 5 M mean age 42 yrs, ﬁve completed and then randomized three to nandrolone and two to megestrol

Interventions a) MA 400 mg/d orally b) nandrolone decanoate 100 mg/fortnight IM injection c) diet counseling duration 12 weeks

Outcomes Weight Appetite VAS ten points score 0 = poor appetite 10 = good appetite Dietary intake %

Notes 12 weeks of treatment QS=2

Risk of bias

Bias Authors’ judgement Support for judgement

Allocation concealment? Unclear risk B - Unclear

Beller 1997

Methods Double blind, randomised controlled multicentre (15), stratiﬁed by institution and whether receiving any anti tumor treatment

Participants 240 cancer pts a) 81 pts = 53M + 28F (nine pts = 50 yrs, 49 pts = 51 to 70 yrs, 23pts = > 71 yrs) b) 80 pts = 52M + 28F (seven pts = lower than 50 yrs, 52 pts = 51 to 70 yrs, 21 pts = >70 yrs) c) 79 pts = 54M + 25F (12 pts = 50 yrs, 51 pts = 51 to 70 yrs, 16 pts = >70 yrs)

Interventions a) MA 480 mg/d orally b) MA 160 mg/d c) placebo

Outcomes QoL ﬁve linear analogue self assessment scales (LASA) asked patients about ﬁve factors: physical well being, mood, pain, nausea and vomiting, and appetite. and a LASA uniscale of overall QoL and the Spitzer QL - Index, completed by the clinician - Appetite (LASA score) Nutritional status - weight - triceps skinfold - mid-arm circumference

Notes Two weeks of treatment QS=4

Risk of bias

Bias Authors’ judgement Support for judgement

Allocation concealment? Low risk A - Adequate

Bruera 1990

Methods Double blind, cross over randomised controlled

Participants 40 advanced cancer pts 31 included pts = 25M + 6F

Interventions a) MA 480 mg/d b) placebo

Outcomes Appetite, subjective energy level, well being depression, and pain were assessed with VAS (0 to 100 mm) Nutritional status, weight, triceps skin fold, arm and calf circumference. Caloric intake, in calories

Notes Seven days ﬁrst phase QS=3

Risk of bias

Bias Authors’ judgement Support for judgement

Allocation concealment? Unclear risk B - Unclear

Methods Cross over double blind randomised controlled

Participants 84 advanced cancer pts 47M + 37F mean age 62 yrs

Interventions a) MA 480 mg/d b) placebo

Outcomes Appetite (VAS) weight triceps skinfold QoL Functional Living Index Cancer (FLIC)

Notes Ten days of treatment QS=4

Risk of bias

Bias Authors’ judgement Support for judgement

Allocation concealment? Unclear risk B - Unclear

Chen 1997

Methods Randomised controlled

Participants 129 cancer pts a) 48 pts = 36M + 12F mean age 50 yrs b) 41 pts = 30M + 11F mean age 52 yrs c) 40 pts = 30M + 10F mean age 51 yrs

Interventions a) MA 160 mg/d b) cisapride 10 mg/d c) placebo

Outcomes Weight Appetite score zero to ﬁve

Notes Four and eight weeks of treatment during a full course of radiotherapy QS=2

Risk of bias

Bias Authors’ judgement Support for judgement

Allocation concealment? Unclear risk B - Unclear

De Conno 1998

Methods Double blind randomised controlled

Participants 42 cancer pts a) 21 pts = 15M + 6F mean age 63 yrs b) 21 pts = 16M + 5F mean age 58 yrs

Interventions a) MA 320 mg/d b) placebo

Outcomes Appetite Score zero to ten Weight QoL Therapy Impact Questionnaire (TIQ)

Notes 14 days of treatment QS=3

Risk of bias

Bias Authors’ judgement Support for judgement

Allocation concealment? Unclear risk B - Unclear

Erkurt 2000

Methods Randomised controlled

Participants 115 cancer pts a) 58 pts = 38M + 12F mean age 51 yrs b) 57 pts = 45M + 5F mean age 57 yrs

Interventions a) MA 480 mg/d b) placebo

Outcomes Weight Appetite score zero to ﬁve QoL Eastern Cooperative Oncology Group Performance status (ECOG PS)

Notes Three months of treatment 46 pts during RT, and four at the end of RT of MA group QS=2

Risk of bias

Bias Authors’ judgement Support for judgement

Allocation concealment? Unclear risk B - Unclear

Eubanks 2002

Methods Double blind randomised controlled

Participants 17 cystic ﬁbrosis pts a) ten pts = 5M + 4F range age six to 18 yrs. 1F 35 yrs b) 7pts = 3F + 4M range age six to 15 yrs

Interventions a) MA 10 mg/kg per day b) placebo

Outcomes Weight Triceps skinfold Mid arm circumference

Notes Six months of treatment QS=4

Risk of bias

Bias Authors’ judgement Support for judgement

Allocation concealment? Low risk A - Adequate

Feliu 1992

Methods Double blind randomised controlled

Participants 150 cancer pts a) 76 pts = 58M + 8F mean age 57 yrs b) 74 pts = 55M + 7F mean age 58 yrs

Interventions a) MA 240 mg/d b) placebo

Outcomes Weight Appetite VAS score QoL performance status KI

Notes Two months of treatment QS=4

Risk of bias

Bias Authors’ judgement Support for judgement

Allocation concealment? Unclear risk B - Unclear

Fietkau 1996

Methods Double blind randomised controlled

Participants 64 cancer pts 61 pts = a) 31pts = 25M + 6F mean age 52 yrs b) 30 pts = 24M + 6F mean age 48 yrs

Interventions a) MA 160 mg/d b) placebo

Outcomes Nutritional status weight QoL index according to Padilla Index

Notes Six weeks of treatment during and up six weeks following radiotherapy Pts were stratiﬁed according oral feeding versus gastrostomy QS=3

Risk of bias

Bias Authors’ judgement Support for judgement

Allocation concealment? Unclear risk B - Unclear

Methods No data available

Participants No data available

Interventions No data available

Outcomes No data available

Notes No data available

Risk of bias

Bias Authors’ judgement Support for judgement

Allocation concealment? Unclear risk D - Not used

Gebbia 1996

Methods Randomised controlled

Participants 122 cancer pts a) 62 pts = 46M + 16F mean age 63 yrs b) 60 pts = 42M + 18F mean age 65 yrs

Interventions a) MA 160 mg/d b) MA 320 mg/d

Outcomes Appetite, Symptom Distress Scale (SDS) Weight Performance Status

Notes 30 days of treatment QS=2

Risk of bias

Bias Authors’ judgement Support for judgement

Allocation concealment? Unclear risk B - Unclear

Methods Randomised controlled

Participants 66 cancer a) 33 pts = 24M + 9F mean age 65 yrs b) 33 pts = 27M + 6F mean age 68 yrs

Interventions a) MA 160 mg/d b) MA 480 mg/d

Outcomes Appetite (subjective ten point scale) Weight

Notes Four weeks of treatment QS=1

Risk of bias

Bias Authors’ judgement Support for judgement

Allocation concealment? Unclear risk B - Unclear

Jatoi 2002

Methods Double blind randomised controlled multicentre (20)

Participants 469 cancer pts a) 159 pts = 103M + 56F mean age 65 yrs b)152 pts = 100M + 52F mean age 67 yrs c) 158 pts = 104M + 54F mean age 67 yrs

Interventions a) MA 800 mg/d liquid suspension +2 capsules placebo b) Dronabinol capsules 2.5 mg orally x 2 + liquid placebo c) MA suspension 800 mg/d + Dronabinol capsules 2.5 mg x 2

Outcomes Appetite (validated questionnaires) Weight QoL Functional Assessment of Anorexia/Cachexia Therapy (FAACT) instrument uniscale and thirteen item

Notes Duration more than four weeks of treatment QS=3

Risk of bias

Bias Authors’ judgement Support for judgement

Allocation concealment? Unclear risk B - Unclear

Jatoi 2004

Methods Double blind randomised controlled multicentre (26)

Participants 421 cancer pts a) 140 pts = 97M + 43F mean age 65 yrs b) 141 pts = 104M + 37F mean age 66 yrs c) 140 pts = 92M + 48F mean age 66 yrs

Interventions a) MA 600 mg/d liquid suspension + isocaloric, isonitrogenous placebo cans b) EPA supplement, two cans/d + placebo liquid suspension c) EPA supplement two cans/d plus MA liquid suspension 600 mg/d orally in combination

Outcomes Weight Appetite (NCCTG questionnaire and FAACT) QoL (single-item Uniscale)

Notes Duration more than three months QS=4

Risk of bias

Bias Authors’ judgement Support for judgement

Allocation concealment? Unclear risk B - Unclear

Lai 1994

Methods Randomised controlled

Participants 52 Cancer pts a) 18 pts = 2M + 16F mean age 64 yrs b) 17 pts = 17F mean age 60 yrs c) 17 pts = 2M + 15F mean age 57 yrs

Interventions a) MA 160 mg/d b) Prednisolone 30 mg/d c) placebo x 3

Outcomes Weight Appetite, self report QoL performance status KI

Notes Duration 21 days QS=2

Risk of bias

Bias Authors’ judgement Support for judgement

Allocation concealment? Unclear risk B - Unclear

Loprinzi 1990b

Methods Double blind randomised controlled

Participants 133 Cancer pts a) 67 pts = 44M + 23F mean age 69 yrs b) 66 pts = 44M + 22F mean age 67 yrs

Interventions a) MA 800 mg/d b) placebo

Outcomes Weight Appetite

Notes One month of treatment QS=5

Risk of bias

Bias Authors’ judgement Support for judgement

Allocation concealment? Low risk A - Adequate

Loprinzi 1994

Methods Randomised controlled

Participants 342 cancer pts a) 88 pts = 56M + 32F mean age 67 yrs b) 86 pts = 54M + 32F mean age 67 yrs c) 85 pts = 55M + 30F mean age 67 yrs

d) 83 pts = 54M + 29F mean age 66 yrs

Interventions a) MA 160 mg/d b) MA 480 mg/d c) MA 800 mg/d d) MA 1280 mg/d

Outcomes Weight Appetite

Notes Median 66 days of treatment QS=3

Risk of bias

Bias Authors’ judgement Support for judgement

Allocation concealment? Low risk A - Adequate

Loprinzi 1999a

Methods Randomised controlled multicentre

Participants 475 cancer pts a) 79 pts = 55M + 30F mean age 66 yrs b) 159 pts = 99M + 60F mean age 66 yrs

Interventions a) MA 800 mg/d b) dexamethasone 3 mg/d

Outcomes Appetite Weight QoL (uniscale)

Notes One month of treatment QS=3

Risk of bias

Bias Authors’ judgement Support for judgement

Allocation concealment? Low risk A - Adequate

Methods Randomised controlled multicentre

Participants 475 cancer pts a) 79 pts = 54M + 29F mean age 66 yrs b) 158 pts = 100M + 58F mean age 67 yrs

Interventions a) MA 800 mg/d b) ﬂuoxymesterone 20 mg/d

Outcomes Appetite Weight QoL (uniscale)

Notes One month of treatment QS=3

Risk of bias

Bias Authors’ judgement Support for judgement

Allocation concealment? Low risk A - Adequate

Marchand 2000

Methods Double blind cross over controlled randomised

Participants 12 cystic ﬁbrosis children pts a) six pts b) six pts mean age seven, four yrs, 9F + 3M

Interventions a) MA 10 mg/ kg/d x 2 b) placebo

Outcomes Appetite Weight Triceps skinfold Mid arm circumference

Notes 12 weeks of treatment QS=3

Risk of bias

Bias Authors’ judgement Support for judgement

Allocation concealment? Unclear risk B - Unclear

McMillan 1994

Methods Randomised controlled

Participants 38 cancer pts a) 20 pts mean age 73 yrs b) 18 pts mean age 70 yrs

Interventions a) MA 480 mg/d b) placebo

Outcomes Weight

Notes 12 weeks of treatment QS=2

Risk of bias

Bias Authors’ judgement Support for judgement

Allocation concealment? Unclear risk B - Unclear

McQuellon 2002

Methods Double blind randomised controlled

Participants 57 cancer pts a) 28 pts = 20M + 8F, mean age 60 yrs b) 28 pts = 16M + 12F mean age 65 yrs (one excluded due to positive ﬁndings in bone scan)

Interventions a) MA 800 mg/ d suspension b) placebo suspension 20 ml

Outcomes Weight QoL FACT

Notes 12 weeks of treatment QS=3

Risk of bias

Bias Authors’ judgement Support for judgement

Allocation concealment? Low risk A - Adequate

Oster 1994

Methods Double blind randomised controlled multicenter (13)

Participants 100 HIV pts a) 52 pts = 50M + 2F mean age 40 yrs b) 48 pts = 47M + 1F mean age 40 yrs

Interventions a) MA 800 mg/d suspension b) placebo, suspension

Outcomes Appetite Weight Triceps skinfold Mid arm circumference Performance status KI

Notes 12 weeks of treatment QS=5

Risk of bias

Bias Authors’ judgement Support for judgement

Allocation concealment? Low risk A - Adequate

Pardo 2003

Methods Randomized controlled

Participants 130 cancer patients a) 64 pts = 58M + 6F mean age 66.8 yrs b) 66 pts = 62 M + 4F mean age 65.3 yrs

Interventions a) MA 320mg/day b) MA 600mg/day

Outcomes Appetite

Notes 30 days of treatment QS=1

Risk of bias

Bias Authors’ judgement Support for judgement

Allocation concealment? Unclear risk B - Unclear

Rowland 1996

Methods Double blind randomised controlled

Participants 243 cancer pts a) 122 pts = 56M + 44F b) 121 pts = 64M + 36F

Interventions a) MA 800 mg/d b) placebo

Outcomes Appetite questionnaire QoL VAS uniscale

Notes Mean duration four months and up to two years QS=2

Risk of bias

Bias Authors’ judgement Support for judgement

Allocation concealment? Unclear risk B - Unclear

Schmoll 1992

Methods Randomised controlled

Participants 91 cancer pts a) 29 pts = 18M + 11F mean age 60 yrs b) 28pts = 16M + 12F mean age 58 yrs c) 34 pts = 25M + 9F mean age 60 yrs

Interventions a) MA 960 mg/d b) placebo c) MA 480 mg/d

Outcomes Appetite Weight

Notes a) median duration eight weeks of treatment b) median duration six weeks c) median duration seven weeks QS=2

Risk of bias

Bias Authors’ judgement Support for judgement

Allocation concealment? Unclear risk B - Unclear

Tchekmedyian 1992

Methods Double blind randomised controlled

Participants 89 cancer pts a) 49 pts = 28M + 9F mean age 63 yrs b) 40 pts = 18M + 12F mean age 64 yrs

Interventions a) MA 1600 mg/d, ten tablets a day, in divided doses b) placebo ten tablets

Outcomes Appetite, categoric scale of ﬁve levels QoL LAS, 29 items Weight Triceps skinfold Mid arm circumference

Notes Six weeks of treatment QS=4

Risk of bias

Bias Authors’ judgement Support for judgement

Allocation concealment? Low risk A - Adequate

Methods Randomised controlled multicenter (9)

Participants 50 HIV pts a) 12 pts = 10M + 2F mean age 46 yrs b) 12 pts = 10M + 2F mean age 39 yrs c) 13 pts = 12M + 1F mean age 38 yrs d) 13 pts = 12M + 1F mean age 40 yrs

Interventions a) MA 750 mg/d, tablets x 1 b) Dronabinol 5 mg/d, tablets x 2 c) a) + b) d) MA 250 mg/d + Dronabinol 5 mg/d, two tablets

Outcomes Weight QoL KI

Notes 12 weeks of treatment QS=2

Risk of bias

Bias Authors’ judgement Support for judgement

Allocation concealment? Unclear risk B - Unclear

Ulutin 2002

Methods Randomised controlled study

Participants 119 cancer pts a) 59 pts = 48M + 11F mean age 56 (range 38 to 72) b) 60 pts = 47M + 13F mean age 58 (range 40 to 74)

Interventions a) MA 160 mg/d orally b) MA 320 mg/d orally in two divided doses 12 hrs apart

Outcomes Weight Appetite (Symptom Distress Scale) QoL (ten point scale) based on patient statements

Notes Three month duration treatment QS=2

Risk of bias

Bias Authors’ judgement Support for judgement

Allocation concealment? Unclear risk B - Unclear

Vadell 1998

Methods Double blind randomised controlled multicenter (nine)

Participants 150 cancer pts a) 49 pts = 38M + 11F mean age 66 yrs b) 51 pts = 42M + 9F mean age 63 yrs c) 50 pts = 31M + 19F mean age 65 yrs

Interventions a) MA 480 mg/d, three tablets b) placebo, three tablets c) MA 160 mg/d, one tablet + placebo two tablets

Outcomes Weight Mid arm circumference Triceps skinfold QoL KI

Notes 12 weeks of treatment QS=3

Risk of bias

Bias Authors’ judgement Support for judgement

Allocation concealment? Unclear risk B - Unclear

Von Roenn 1994

Methods Double blind randomised controlled multicenter

Participants 270 HIV pts a) 75 pts = 75M mean age 38 yrs b) 75 pts = 75M mean age 39 yrs c) 82 pts = 81M + 1F mean age 39 yrs d) 38 pts = 38M mean age 38 yrs

Interventions a) MA 800 mg/d, suspension b) MA 400 mg/d, suspension c) MA 100 mg/ d, suspension d) placebo, suspension

Outcomes Weight Appetite Mid arm circumference Triceps skinfold QoL, by Linear Analog Self Assessment questionnaire

Notes 12 weeks of treatment QS=4

Risk of bias

Bias Authors’ judgement Support for judgement

Allocation concealment? Low risk A - Adequate

Weisberg 2002

Methods Double blind randomised controlled multicenter (18)

Participants 145 COPD pts a) 72 pts = 46M + 26F mean age 68 yrs b) 73 pts = 45M + 28F mean age 66 yrs

Interventions a) MA 800 mg/d, suspension b) placebo, suspension

Outcomes Weight Triceps skinfold Mid arm circumference Appetite

Notes Eight weeks of treatment QS=4

Risk of bias

Bias Authors’ judgement Support for judgement

Allocation concealment? Unclear risk B - Unclear

Methods Double blind randomised controlled multicenter (15)

Participants 255 cancer pts a) 128 pts = 77M + 51F mean age 71 yrs b) 127 pts = 64M + 63F mean age 69 yrs

Interventions a) MA 320 mg/d, tablets b) placebo, tablets

Outcomes Weight QoL EORTC QLQ-C30 Appetite

Notes 12 weeks of treatment QS=5

Risk of bias

Bias Authors’ judgement Support for judgement

Allocation concealment? Low risk A - Adequate

Yeh 2000

Methods Double blind randomised controlled

Participants 69 geriatric pts a) 36 pts = 35M + 1F mean age 76 yrs b) 33 pts = 31M + 2F mean age 76 yrs

Interventions a) MA 800 mg/d, suspension b) placebo, suspension

Outcomes Weight Appetite QoL VAS nine items

Notes 12 weeks of treatment QS=5

Risk of bias

Bias Authors’ judgement Support for judgement

Allocation concealment? Low risk A - Adequate

Methods Double blind randomized controlled

Participants 33 cancer pts a) 16 pts (no data available of age and sex) b) 17 pts b) 17 pts

Interventions a) 320 mg/day, tablets b) placebo

Outcomes Appetite measured in a categoric numeric Scale Weight Performance status (Karnofsky)

Notes A) phase A 14 days of treatment B) phase B, open of 76 days of treatment QS=2

Risk of bias

Bias Authors’ judgement Support for judgement

Allocation concealment? Unclear risk B - Unclear

Pts = patients = females (F) + males (M) VAS = Visual Analog Self reported scale *** = CHRONIC OBSTRUCCION PULMONARY DISEASE (COPD) KI = Karnofsky Index yrs = Years M = male F = female MA - Megesterol acetate QS = Quality Score (Oxford Quality Scale)

Characteristics of excluded studies [ordered by study ID]

Study Reason for exclusion

Macbeth 1994 This study was unpublished and closed early due to the numbers of withdrawals (> 50% of death in the megestrol arm)

McMillan 1999 This study was excluded because the two arms of the treated patients were receiving megestrol acetate

Yeh 2004 This study involved the same participants as Yeah 2000, in which they measured different outcomes

Comparison 1. Megestrol acetate vs placebo (ITT)

No. of No. of Outcome or subgroup title studies participants Statistical method Effect size

1 Appetite improvement 10 885 Risk Ratio (M-H, Random, 95% CI) 2.76 [1.65, 4.61] 1.1 Cancer 8 565 Risk Ratio (M-H, Random, 95% CI) 3.03 [1.83, 5.01] 1.2 AIDS 1 308 Risk Ratio (M-H, Random, 95% CI) 1.34 [0.97, 1.86] 1.3 Other underlying 1 12 Risk Ratio (M-H, Random, 95% CI) 13.0 [0.89, 189.39] pathology 2 Weight gain 10 843 Mean Difference (IV, Random, 95% CI) 2.78 [1.61, 3.95] 2.1 Cancer 6 499 Mean Difference (IV, Random, 95% CI) 3.56 [1.27, 5.85] 2.2 AIDS 1 113 Mean Difference (IV, Random, 95% CI) 4.26 [2.93, 5.59] 2.3 Other underlying 3 231 Mean Difference (IV, Random, 95% CI) 0.65 [-0.14, 1.44] pathology 3 Weight gain 10 1246 Risk Ratio (M-H, Random, 95% CI) 1.94 [1.43, 2.63] 3.1 Cancer 7 895 Risk Ratio (M-H, Random, 95% CI) 2.14 [1.41, 3.24] 3.2 AIDS 1 270 Risk Ratio (M-H, Random, 95% CI) 2.60 [1.39, 4.87] 3.3 Other underlying 2 81 Risk Ratio (M-H, Random, 95% CI) 1.35 [0.83, 2.21] pathology 4 Quality of life 6 952 Risk Ratio (M-H, Random, 95% CI) 1.52 [0.87, 2.66] 4.1 Cancer 4 582 Risk Ratio (M-H, Random, 95% CI) 1.52 [0.79, 2.91] 4.2 AIDS 2 370 Risk Ratio (M-H, Random, 95% CI) 1.49 [0.47, 4.69] 4.3 Other underlying 0 0 Risk Ratio (M-H, Random, 95% CI) 0.0 [0.0, 0.0] pathology 5 Side effects 11 1767 Risk Ratio (M-H, Random, 95% CI) 1.74 [1.29, 2.35] 5.1 Edema 11 1767 Risk Ratio (M-H, Random, 95% CI) 1.74 [1.29, 2.35]

Comparison 2. Megestrol acetate vs other drugs (ITT)

No. of No. of Outcome or subgroup title studies participants Statistical method Effect size

1 Appetite improvement 3 514 Risk Ratio (M-H, Random, 95% CI) 1.15 [0.75, 1.76] 1.1 Cancer 3 514 Risk Ratio (M-H, Random, 95% CI) 1.15 [0.75, 1.76] 1.2 AIDS 0 0 Risk Ratio (M-H, Random, 95% CI) 0.0 [0.0, 0.0] 1.3 Other underlying 0 0 Risk Ratio (M-H, Random, 95% CI) 0.0 [0.0, 0.0] pathology 2 Weight gain 5 598 Mean Difference (IV, Random, 95% CI) 3.82 [0.06, 7.57] 2.1 Cancer 3 564 Mean Difference (IV, Random, 95% CI) 1.53 [-0.18, 3.25] 2.2 AIDS 2 34 Mean Difference (IV, Random, 95% CI) 8.40 [7.46, 9.35] 2.3 Other underlying 0 0 Mean Difference (IV, Random, 95% CI) 0.0 [0.0, 0.0] pathology 3 Weight gain 4 524 Risk Ratio (M-H, Random, 95% CI) 1.30 [0.88, 1.90] 3.1 Cancer 3 514 Risk Ratio (M-H, Random, 95% CI) 1.60 [0.85, 3.03]