GeneSight (1): A Brief Critical Read on the Impact of pharmacogenomics on clinical outcomes in major depressive disorder in the GUIDED trial: A large, patient- and rater-blinded, randomized, controlled study (2)
Gene Simmons, Gene&Tonic, GeneSight (1): What’s all the Hype?
BLUF (bottom line up front): Study has limitations that limit the usefulness of the Genesight testing in the treatment of depression. Limitations:
- Small number of study participants
- Short evaluation period
- Null hypothesis on primary outcome was true
- Flawed assumption
Major depressive disorder (MDD) is a treatable mood disorder with an estimated 15.7 million adults aged 18 or older in the United States having at least one major depressive episode in the past year (3). This number represents a prevalence of 7% of all adults in the US population (3,4). That being said, the referenced study here (1) called itself a “A large, patient- and rater-blinded, randomized, controlled study (1).” The study had an N of 1398 with 717 in the treatment as usual (control) arm, and 681 in the study arm. Ok, if there are 15.7 million adults with depression in the USA and this study had only 1398 of them, could we say this study is underpowered? This is a study population of .08% of the target population. Compare this to the iconic The Sequenced Treatment Alternatives to Relieve Depression (STAR*D) (5) which had an N of 2876 or more than double the Genesight study. I’m not passing judgment, but just reporting the facts. Make your own judgment. In the past 2 weeks, 2 patients have brought up the Genesight testing and I honestly didn’t know much about it, so to be prudent, I reviewed the data. The first limitation I see however is the low number of participants in the study.
“Pharmacogenomic testing did not significantly improve mean symptoms but did significantly improve response and remission rates for difficult-to-treat depression patients over standard of care”
Holy Null Hypothesis. That’s twisted …….. sister!
Need I say more? The actual hypothesis isn’t explicitly stated but can be inferred as something to the effect: Pharmacogenomic testing will lead to improved symptoms when compared to treatment as usual in patients with MDD. The primary outcome was symptom improvement at week 8 in the per-protocol cohort. However, at week 8, there was no significant difference in the primary outcome: “The primary outcome of symptom improvement was not significantly different between the study arms.” 8 weeks to me doesn’t seem like enough time. The study did appropriately report the lack of statistical significance in the primary outcome. However it went on to report positive secondary outcomes. Sort of a bait and switch: These results from the GUIDED trial indicate that pharmacogenomics testing is effective in improving response and remission rates among those with prior treatment resistance, particularly for patients who are treated with medications that are incongruent with their genetic profile.
Lastly, there is a major flawed assumption. The assumption is that the level of drug in the body, as a function of hepatic metabolism, can predict clinical response. The GeneSight® Psychotropic test from Assurex Health, Inc. (Mason, OH) was used for pharmacogenomic testing on all patients. This measured the genotypes of 59 alleles and variants across 8 genes (CYP1A2, CYP2C9, CYP2C19, CYP3A4, CYP2B6, CYP2D6, HTR2A, SLC6A4). A genotype is developed for each person. Thus the genotype of how a person will metabolize drugs is then compared to how the actual drugs are metabolized and a prediction of clinical response is made. In other words, a person per genotype might be a heavy 2D6 metabolizer. Fluoxetine which is a 2D6 substrate – the prediction is that the person will not respond to fluoxetine. However bupropion, which is not a 2D6 substrate – the prediction is that they will respond. This was called a gene-drug interaction and 38 psychotropic medications were categorized based on three levels of gene-drug interaction. They are saying that the metabolism of the drugs based on genotype can be used to predict clinical response. This is quite a leap. There were other items in the study to be addressed, but this is enough for now. I have the gist of the trial and can now better guide patients when they bring it up again. Gene & Tonic anyone?
- Greden JF, Parikh SV, Rothschild AJ, Thase ME, Dunlop BW, DeBattista C, et al. Impact of pharmacogenomics on clinical outcomes in major depressive disorder in the GUIDED trial: a large, patient-and rater-blinded, randomized, controlled study. J Psychiatr Res. (2019) 111:59–67. doi: 10.1016/j.jpsychires.2019.01.003
- Center for Behavioral Health Statistics and Quality (2015) Behavioral health trends in the United States: Results from the 2014 National Survey on Drug Use and Health.
- Zelazny K, Simms LJ (2015) Confirmatory factor analyses of DSM-5 posttraumatic stress disorder symptoms in psychiatric samplesmdiffering in Criterion A status. J Anxiety Disord 34: 15-23.
- Jablonski, Michael & King, Nina & Wang, Yongbao & Winner, Joel & Watterson, Lucas & Gunselman, Sandra & Dechairo, Bryan. (2018). Analytical validation of a psychiatric pharmacogenomic test. Personalized Medicine. 15. 10.2217/pme-2017-0094. https://www.futuremedicine.com/doi/10.2217/pme-2017-0094?url_ver=Z39.88-2003&rfr_id=ori:rid:crossref.org&rfr_dat=cr_pub%3dwww.ncbi.nlm.nih.gov