Late-Stage Drug Development

Late-Stage Drug Development 2017-11-01T14:09:55+00:00


In an alliance with Takeda Pharmaceutical Company Limited, our consultants working within Zinfandel wrote the clinical protocol for the ongoing TOMMORROW study. This is a Phase III registration trial to evaluate an algorithm comprising the genotypes APOE (apolipoprotein E) and TOMM40 (translocase of outer mitochondrial membrane) together with age as an indicator of risk for Alzheimer’s disease. This trial is also designed to test the efficacy of pioglitazone in delaying cognitive impairment due to Alzheimer’s in cognitively normal but high-risk individuals. The alliance developed and validated neuropsychological assessment tools, operationalized a novel clinical diagnosis, and reached agreement with US and EU regulatory authorities on study endpoints. About 3500 subjects have been recruited at 58 sites in 5 countries on 3 continents.

Using genetics to design an Alzheimer’s study: The TOMMORROW study is designed to qualify a risk-assessment algorithm and to test the potential of the drug pioglitazone in delaying the onset of mild cognitive impairment in cognitively normal but high-risk individuals. All study subjects are asymptomatic at study entry. They are divided into low- or high-risk groups according to the algorithm, which consists of a composite genetic biomarker and age. The biomarker comprises varying APOE4 and TOMM40 genotypes. Subjects deemed to be at high-risk of impairment during their five years in the trial are then randomly assigned to treatment or placebo.


In collaboration with Eli Lilly and Company, a Cabernet team incorporated genetic testing into clinical trials of prasugrel, a P2Y12-receptor inhibitor that inhibits platelet activity. P2Y12-receptor inhibitors, used in combination with aspirin as antiplatelet therapy for coronary-artery disease, are prodrugs that require conversion to active metabolites via the cytochrome P450 (CYP) system. The PGx issue was that the enzymatic activity of a different P2Y12-receptor inhibitor, clopidogrel, was reduced among patients carrying certain “reduced-function” alleles of CYP2C19. These patients exhibited higher platelet reactivity than did clopidogrel-treated patients without reduced-function alleles and appeared to have greater risk of ischemic events after percutaneous coronary intervention. Accordingly, the Food and Drug Administration added to the prescribing information for clopidogrel a boxed warning that indicated a role for genetic testing. The agency also asked Lilly to evaluate the potential CYP2C19 genetic liability of prasugrel. Our consultants coordinated the genotyping; the analysis of pharmacokinetics, pharmacodynamics, and clinical efficacy; and the regulatory reporting. Several clinical studies demonstrated that CYP2C19 genotype did not appear to be associated with platelet reactivity or clinical outcomes for patients treated with prasugrel, in contrast to clopidogrel. Genotyping could therefore guide selection of prasugrel versus clopidogrel. A point-of-care medical device for determining CYP2C19 genotype is now commercially available. Braun et al, 2013, Erlinge, et al, 2014, Gurbel et al, 2014, Mega et al, 2015, Doll et al, 2016

The prasugrel work was one part of a wide-ranging, multi-year collaboration with Lilly. The opioid-receptor antagonist and evacetrapib entries under Early-Stage Drug Development provide other examples. (Cabernet-Lilly Alliance, 2008, Akkari et al, 2009, Burns & Sundseth, 2012, Swanson et al, 2014, Bishop et al, 2015)

Infographics_text outline-10
Distinguishing between antiplatelet therapies: Prasugrel and clopidogrel both are prodrugs that require conversion by the cytochrome P450 (CYP) system to active metabolites that then serve as therapy for coronary-artery disease. But genetic testing demonstrated that in a definable population this conversion differs between the two prodrugs. They thus differ in their capacity to inhibit P2Y12 platelet receptors and reduce clotting. Specifically, conversion of clopidogrel was significantly reduced among patients carrying “reduced function” alleles of CYP2C19 (Figure A). These “reduced metabolizers,” in contrast to “extensive metabolizers,” showed significantly greater platelet reactivity when treated with clopidogrel but not when treated with prasugrel (Figures B and C). Reactivity was measured with two standard assays (VerifyNow PRU in B, VASP PRI in C).


While on staff at GlaxoSmithKline, scientists now at Cabernet helped to design PGx analyses to address a hypersensitivity reaction experienced by a small percentage of patients repeatedly treated with the HIV/AIDS drug abacavir, a nucleoside reverse transcriptase inhibitor. The risk resulted in a black-box warning and a post-marketing pharmacovigilance program. The PGx analyses identified a marker, HLA-B*5701, of highly specific and sensitive clinical utility. On the strength of these data, regulatory authorities in the US and the EU recommended screening for the HLA-B*5701 allele before initiating abacavir therapy. Even before the US labeling change, use of the diagnostic test for HLA-B*5701 increased nine-fold in the six months following the announcement of the PGx-study results. Roses et al, 2007, Mallal et al, 2008

Preventing a drug’s toxic effect: Although the HIV/AIDS drug abacavir is generally well tolerated, a fraction of patients experience hypersensitivity reactions, and re-treating with the drug following a reaction can be life-threatening. A way to refine therapy and improve safety became possible when genetic analysis identified a marker, HLA-B*5701, which closely correlated with hypersensitivity. The clinical utility of this discovery was demonstrated in a global study that randomized patients to either the then prevailing standard of care, which did not include screening for HLA-B*5701, or to care that did include screening. Patients in the screened group who were found to be HLA-B*5701-negative progressed to abacavir therapy; patients who were HLA-B*5701-positive received a different HIV/AIDS therapy. Among the unscreened, standard-of-care group, 7.8% were clinically suspected to have developed hypersensitivity reactions following abacavir therapy, and 2.7% were immunologically confirmed as hypersensitive by means of a skin-patch test. Among the screened, HLA-B*5701-negative patients treated with abacavir, only 3.4% were clinically suspected to have hypersensitivity reactions, and none of these suspected cases could be confirmed with the skin-patch test.