Did Betrixaban Really Fail in the APEX Study?

Did the drug fail that bad?

Portola Pharmaceuticals announced results on March 24, 2016, from the phase 3 APEX trial, which sought to evaluate superiority of a new direct factor Xa inhibitor (betrixaban) against enoxaparin (Lovenox) for the prevention of venous thromboembolism (VTE) in acute medically ill patients.1

Shortly after the results were announced, stock price plummeted 30%. Did the drug fail that bad? Let’s take a closer look at the study.

What happened?

Multiple headlines suggested the study failed to find a significant difference, but the study results looked a little different this time. Conventionally, there would be only the comparison between the treatment group (betrixaban) and control group (enoxaparin).

This study reported results from 2 treatment cohorts of betrixaban (with the same dose, but different population) against enoxaparin (Table 1).1

Table 1. Summary of Results from APEX Study1

Primary efficacy analysis (VTE)

Primary safety analysis (major bleeding)†


P value


P value

Cohort 1: Elevated D-Dimer





Cohort 2: Elevated D-dimer or age ≥75 years





Overall study population





*A p value of <0.05 is accepted as the gold standard for statistical significance.

†No significant difference in major bleeding between any groups.

What confused me was that everyone was fixated on cohort 1, not the overall study population.

Again, conventionally you would look toward the overall study population for clinical significance, not a subpopulation (such as in this study).

What piqued my interest was that the press release also mentioned that APEX was the first thrombosis prevention study to incorporate an enrichment strategy from an FDA guidance.

What does enrichment of a study population mean?

The FDA defines enrichment as “the prospective use of any patient characteristic to select a study population in which detection of a drug effect (if one is in fact present) is more likely than it would be in an unselected population.”

In short, the FDA wanted to give some advice on how to best select a population to demonstrate efficacy.2 Table 2 gives a brief outline of those strategies.

Table 2. Strategies for Study Design Enrichment2

Main Strategy Categories



Decrease heterogeneity

Selecting patients with baseline measurements in a narrow range. Excluding patients whose disease may be highly variable.

Restrictive inclusion and exclusion criteria of a study to evaluate a very specific population.

Prognostic enrichment

Selecting patients with a high likelihood of having a disease-related endpoint.

Evaluating patients with a high risk for cardiovascular events in a statin study.

Predictive enrichment

Selecting patients with a high likelihood of treatment response, based on patient’s physiology or disease characteristic.

Use of markers (eg, tumor EGFR) to identify potential responders to chemotherapy.

Where did the phase 3 APEX study go wrong?

The initial study inclusion and exclusion criteria that was published had a composite primary outcome and broader inclusion criteria from what was presented in their press release.

What Portola might have done was select a population (enriched population) that they thought would have responded the best, plus a single outcome (VTE) they thought would provide the most pronounced effects. In this case, this was cohort 1 and cohort 2. Both an elevated D-Dimer and age ≥75 years were considered major risk factors for a VTE. However, this strategy may have backfired when a non-statistically significant result was found in cohort 1.3,4

Would it have been better to just use the entire population?

In this case, probably yes. Traditionally the entire study population is evaluated for efficacy first, before subgroups are evaluated to see if there is a pronounced effect in a subpopulation.

In this study, a smaller enriched population was evaluated first as the main study population for approval.

While an enriched population might provide a more precise and useful target population, it was fatal in this instance. If the study results were flipped around, headlines could have read: “Betrixaban superior to enoxaparin for VTE prophylaxis in the medically ill! May be more beneficial in older patients.”

Will betrixaban be approved based on trial results?

The entire study has yet to be published, presented at any conference, or submitted to the FDA. Between now and then, additional study results may be presented to try and bolster the treatment effect of betrixaban.

Things that count against the approval of betrixaban are other factor Xa inhibitors that are on the market (apixaban, dabigatran, edoxaban, rivaroxaban). However, none of these agents are approved for the prevention of VTE in medically ill patients, even though some have tried (eg, apixaban).5

The draft guidance did note that the FDA was “prepared to approve drugs studied primarily or even solely in enriched populations … but the extent of data that should be available on the non-enriched subgroups should always be considered.”

While there was no superiority demonstrated in cohort 1, the FDA may greenlight the approval of betrixaban for medically ill patients for the prevention of VTE with labeling that suggests the benefit is only observed in certain populations.

Take-away points

· Interesting presentation of initial study results using enriched populations

· A good, but not definite, chance that betrixaban may be FDA-approved for VTE prophylaxis in medically ill patients (statistically significant 24% relative reduction of VTE compared with enoxaparin in overall population)

· A difference of 0.004 in a p value can mean the difference between raging success or a 30% loss of stock value


  • Portola Pharmaceuticals Announces Topline Results from Phase 3 APEX Trial of Betrixaban for Prevention of Blood Clots in Acute Medically Ill Patients. Portola website. http://investors.portola.com/phoenix.zhtml?c=198136&p=irol-newsroomArticle&ID=2150685. Accessed March 24, 2016.
  • Guidance for Industry Enrichment Strategies for Clinical Trials to Support Approval of Human Drugs and Biological Products. FDA website. http://www.fda.gov/downloads/Drugs/GuidanceComplianceRegulatoryInformation/Guidances/UCM332181.pdf. Accessed March 24, 2016.
  • Goldhaber SZ, Leizorovicz A, Kakkar AK, et al. Apixaban versus enoxaparin for thromboprophylaxis in medically ill patients. N Engl J Med. 2011;365(23):2167-2177.
  • Cohen AT, Harrington R, Goldhaber SZ, et al. The design and rationale for the Acute Medically Ill Venous Thromboembolism Prevention with Extended Duration Betrixaban (APEX) study. Am Heart J. 2014;167(3):335-341.