A phase III randomized study comparing the effects of oxandrolone (Ox) and megastrol acetate (Meg) on lean body mass (LBM), weight (wt) and quality of life (QOL) in patients with solid tumors and weight loss receiving chemotherapy

Reviewer: Arpi Thukral, MD
Abramson Cancer Center of the University of Pennsylvania
Ultima Vez Modificado: 3 de junio de 2008

Presenter: G. J. Lesser
Presenter's Affiliation: Wake Forest University, Winston-Salem, NC
Type of Session: Scientific

  • A study by DeWys, et. al. in 1980 showed that unintentional weight loss in cancer patients is a significant problem and is a poor prognostic factor. It is associated with decreased survival and QOL, and increased toxicity to cancer treatment.
  • The weight loss in cancer patients is disproportionately seen as a loss of muscle mass rather than fat, which may be a reason for the negative effects mentioned above.
  • Common drugs such as megastrol acetate (or Megace) and corticosteroids (such as dexamethasone and prednisone) have been standardly used to reduce unintentional weight loss, and can actually help patients gain weight.
  • Oxandrolone (Oxandrin, Ox) is an oral anabolic steroid which has a high anabolic-to-androgenic ratio.
    • For this reason, it is proposed that this drug promotes weight gain by promoting muscle regrowth and increasing LBM.
    • Ox has been used for a variety of medical disorders causing involuntary weight loss, and has often been used by body builders to build muscle.
  • The purpose of this phase III randomized study was to compare the effects of Ox to Meg (a current standard) on LBM and weight of patients.
Materials and Methods
  • Cancer patients with solid tumors undergoing chemotherapy who have a >5% of total body weight loss in previous 6 months, or a >3% weight loss in the previous month, were randomized to receive either Ox (100 mg bid) or Meg (800 mg daily) while undergoing chemotherapy.
  • Stratification was by Stage I-III vs. IV, concurrent XRT vs. none, and male vs. female.
  • Eligibility criteria:
    • Adults >18 years
    • Patients with a histologic diagnosis of cancer
    • Patients who have already begun chemotherapy
    • >5% of total body weight loss in previous 6 months
    • >3% weight loss in previous month
    • Concurrent XRT was allowed
    • ECOG Performance status 0-2
    • Life expectancy > 6 months
    • Cr­­<2.5
  • Ineligibility criteria:
    • Patient were not allowed to be on other steroids/hormones, however steroids for anti-emetic use during chemotherapy were allowed.
    • No use of the study drugs within 3 months prior to enrollment.
    • Hormonally responsive tumors such as breast, prostate, and germ cell tumors were excluded.
    • Patients with ascites or pleural effusions were excluded since weight loss data calculation would be difficult.
    • Patients who have had a recent MI.
  • The primary endpoint studied was LBM after 12 weeks of treatment.
  • Secondary endpoints evaluated included weight, fat mass, and QOL.
  • Serial assessments of these endpoints were performed at 4 weeks, 8 weeks, 12 weeks, and one month post treatment.
  • Body composition was evaluated serially using a Bioelectrical Impedance Analysis, and QOL was evaluated using the FACT-G and the Anorexia/Cachexia subscale.
  • Between 6/2004 and 5/2007, 155 patients were accrued onto the study, and patients were randomized to receive Ox (n=80) or Meg (n=75).
  • In terms of patient characteristics, the groups were fairly balanced.
    • It is notable that 60% of all patients enrolled had advanced stage IV disease.
    • 20% of patients were African American and 80% were Caucasian.
    • 41% of patients were female.
    • 50% of patients had lung cancer and 12.5-20% had colorectal cancer.
    • Median age was 64 years.
  • All data shown was for assessments performed at 12 weeks, and was reported as change from baseline.



p -value

Change in LBM in lbs.




Change in weight in lbs.




Change in fat mass




Change in QOL


Change in self-reported anorexia




Changes in fatigue




  • At 12 weeks, significant changes from baseline were seen for weight and fat mass, as seen in the table above.
  • Although an increase in LBM was seen in both groups, they did not differ significantly. QOL was not impacted by either treatment. There were improvements seen in fatigue, QOL, and anorexia but they were all non-significant.
  • There were a total of 12/96 serious adverse events which were though to be related to the drug.



Related to drug





Increased LFTs




DVT, PE, or thrombus

  • There was a 50% overall dropout rate at 12 weeks, and this was worse in the Ox group (063%) than the Meg group (39%).
Author's Conclusions
  • While patients treated with Ox did lose weight, they also experienced an increase in LBM.
  • A reduction in fat mass and reduced self-reported anorectic symptoms were also seen.
  • Patients treated with Meg had an increase in weight and fat mass, increased appetite, and minimal change on LBM.
  • The authors conclude that these 2 drugs were shown to have complementary effects in terms of overall weight gain and LBM.
  • They suggest that future studies to explore these 2 agents in combination in a similar patient population may be fruitful.
    • They have designed a trial which they would like to implement studying Meg vs. Ox+Meg.
Clinical/Scientific Implications
  • Unintentional weight loss is a significant problem in patients with cancer, and can lead to a lower tolerance to chemotherapy and increased morbidity.
    • Current therapies to improve weight loss such as steroids and progestational agents have only been fairly effective in increasing muscle mass or LBM in these patients.
  • The authors have executed a randomized trial to test a therapy, Ox, which is proposed to increase muscle mass, rather than fat mass.
  • While their results have showed an increase in LBM with Ox, the authors were unable to show a statistically significant difference in change in LBM from that shown by Meg. In addition, this study showed that patients on Ox actually lost weight and fat mass, instead of gaining them. In contrast, patients on Meg did gain weight, even though a greater proportion was likely secondary to fat mass. These results were inconclusive, and do not support the theory that Ox is a better suited therapy for weight gain in cancer patients than Meg.
  • Limitations:
    • Over 60% of patients were Stage IV, and may have had weight loss despite these treatments due to the severity of their disease. The reason for this skewed patient population may be that oncologists preferably prescribe these drugs to patients who have severe weight loss or cachexia, rather than those with less severe weight loss.
    • There was a 50% overall dropout rate at 12 weeks, which was worse in the Ox group than the Meg group, and reasons for why this was the case have not been reported. It may due to increased toxicity with Ox, but this needs to be further explored.
    • This study tested drugs from different classes with completely different mechanisms against each other. It may have been more prudent to compare Ox versus other steroids that are in use, and do have as high an anabolic/androgenic ratio.
    • The population tested was small.
    • Since it has been shown that unintentional weight loss leads to increased morbidity and mortality in previous studies, it may have been wise to also explore survival advantages.
  • The authors have made a good attempt to study a drug shown in previous studies to increase LBM; however, this data should not be used to change current practices.
  • This data is useful to generate future hypotheses and provides a framework for further studies exploring this class of drugs.
  • It may be true that Ox and Meg may work with complementary effects, and it will be useful to see results from this groups future trial studying these two drugs in combination.

View Patient Summary

Manténgase informado con las últimas informaciones de OncoLink!   Suscribirse a OncoLink eNews
Ver nuestros archivos de boletines