Oops, you're using an old version of your browser so some of the features on this page may not be displaying properly.

MINIMAL Requirements: Google Chrome 24+Mozilla Firefox 20+Internet Explorer 11Opera 15–18Apple Safari 7SeaMonkey 2.15-2.23

20 Jun 2016
Palliative and supportive care
Bishal Gyawali
Bishal Gyawali

A number of agents to combat cancer cachexia are under investigation and development, of which anamorelin has so-far been the most promising. Recently, the results of two phase 3 trials of anamorelin (ROMANA 1 and 2) have been published.1 Although the authors conclude positively and the accompanying editorial also hails this study as offering hopes,2 the results of the ROMANA trials are actually negative. Here, I caution against the positive interpretation of the trial results based on several rationales.

The ROMANA 1 and 2 trials tested anamorelin in non-small-cell lung cancer patients with cachexia by randomizing 484 and 495 patients respectively 2:1 to receive either anamorelin or placebo. The co-primary endpoints were change in lean body mass (LBM), as an indicator for muscle mass, and handgrip strength, as an indicator for muscle function, over 12 weeks. In both ROMANA 1 and 2, patients receiving anamorelin gained LBM versus placebo (0.99kg v -0.47 kg in ROMANA 1, 0.65kg v -0.98 kg in ROMANA 2) which was statistically significant. However, there was no difference in handgrip strengths between anamorelin versus placebo cohorts in both the studies.

Statistical significance does not imply the efficacy of the drug; judgment is required to assess whether there is any clinical value to the achieved significance.3 In these two trials, the actual increment in LBM over 12 weeks is small: a treatment difference of around 1.5 kgs (+990 and +650 grams respectively with anamorelin versus -470 and -980 grams respectively with placebo in ROMANA 1 and 2). This small improvement, although statistically significant, doesn’t have much clinical impact. Considering that confounders such as nutrition, social support, financial status, exercise status etc. were not controlled, and there was no improvement in muscle power as measured by hand-grip strength, this study should be taken as negative. 

Surely, a cancer patient would not care whether he has a few extra grams of muscle if s/he cannot open the door or take a walk anyway. Furthermore, it cannot be ruled out that the statistical significance achieved in this study could simply be a result of an over-powered trial. Also, as the authors point out, this increment in LBM is not due to increment in skeletal muscle mass alone: some part of this increment in LBM could be due to ascites, effusion or increase tumour burden. Because the tumour response was not measured, the chances that the differences in LBM observed could be due to the differences in cancer burden, and not the drug, cannot be ruled out.

Improving LBM in cancer patients is meaningful if it improves survival or QoL. In ROMANA1 and 2, neither OS nor muscle function – not even fatigue – improved with anamorelin, despite improvement in LBM. Thus, the increase in LBM found in ROMANA trials are just numbers – if it doesn’t translate to benefits in terms of survival or functional ability of patients, the increase in these numbers makes no sense no matter how large.

The only positive outcome was a statistically significant improvement in anorexia-cachexia scale score; how much of that is clinically meaningful remains to be known given that food intake data were not collected. Also, although anamorelin was found to be relatively safe in these trials, long term safety data are unavailable. In fact, there have been some concerns about anamorelin potentially stimulating tumour growth by increasing concentrations of growth hormone and insulin-like growth factor-1.4 A phase 3 trial powered for survival can only answer this question with confidence. Until then, primum non nocere dictates extreme caution in the use of this agent.

Thus, even without considering the cost – which in all likelihood is going to be lofty – anamorelin doesn’t seem to actually benefit patients with cancer. There is no doubt that cancer cachexia represents an area of unmet need with no drug despite years of research and discoveries. However, what the patients need is not just a drug, but a drug that works and is safe. Anamorelin still has doubts on both the fronts.

Finally, I noticed that the study sponsor was involved not only in study design and provision of study materials, but also in data collection, interpretation and writing of the report. Also, the accompanying editorial was commissioned from someone with COI with the same industry that sponsored the study.2 This reminded me of a recent nostalgic editorial by Ioannidis5 that explores the history and intricacies of evidence-based medicine (EBM) and conflicts of interest in academic papers. I request all my fellow young oncologists to go through that editorial once to get a better understanding of how EBM is now, versus what it was supposed to be.


  1. Temel JS, Abernethy AP, Currow DC, et al. Anamorelin in patients with non-small-cell lung cancer and cachexia (ROMANA 1 and ROMANA 2): results from two randomised, double-blind, phase 3 trials. The Lancet Oncology 2016;17:519-31.
  2. Muscaritoli M. Targeting cancer cachexia: we're on the way. The Lancet Oncology 2016.
  3. Chavalarias D, Wallach JD, Li AH, Ioannidis JP. Evolution of Reporting P Values in the Biomedical Literature, 1990-2015. Jama 2016;315:1141-8.
  4. Zhang H, Garcia JM. Anamorelin hydrochloride for the treatment of cancer-anorexia-cachexia in NSCLC. Expert opinion on pharmacotherapy 2015;16:1245-53.
  5. Ioannidis JP. Evidence-based medicine has been hijacked: a report to David Sackett. Journal of clinical epidemiology 2016.

Discussion questions

  1. How clinically relevant is it to improve muscle mass without improving muscle function in cancer patients?
  2. How can we decide whether a statistically significant finding is clinically relevant or not?
  3. How can the academic sector and industry collaborate smoothly for efficient yet trustworthy clinical trial results and reporting?
Last update: 20 Jun 2016

The author has no actual, potential, real or apparent interest to declare and has no involvement that might raise the question of bias in the work reported or in the conclusions, implications, or opinions stated.

This site uses cookies. Some of these cookies are essential, while others help us improve your experience by providing insights into how the site is being used.

For more detailed information on the cookies we use, please check our Privacy Policy.

Customise settings