Pharmaceutical companies are increasingly relying on biomarkers to deliver precision medicine in immuno-oncology. Biomarkers can accelerate drug development and reduce the overall cost; they also allow sponsors to identify failed treatments sooner so that resources are not wasted on expensive, late-stage trials with unsafe or inactive compounds. Finally, these tests lead to better outcomes for patients, which help companies make a stronger case for reimbursement.
However, biomarker discovery requires substantial time and resources. While expenses will likely be outweighed by increased development efficiency, companies must ensure that drug and diagnostic timelines are closely aligned so that the treatment and test can launch simultaneously. Technical, workflow and commercial factors are critical to the successful use of immuno-oncology biomarkers.
What makes a successful biomarker?
Biomarkers in drug development fall into five categories:
- Determine whether a drug hits its target and measure the impact on the pathway being modulated. These biomarkers help developers evaluate the treatment’s mechanism of action and define a biologically effective dose.
- Identify patients who are most likely to respond to the drug or least likely to suffer an adverse event. These tests can become companion diagnostics, which are required prior to treatment, or complementary diagnostics, which are run only to inform the physician.
- Identify patients who could be resistant or become resistant to the drug. For example, mutation analyses can reveal whether a patient has a genetic abnormality linked with resistance.
- Predict the course of disease independent of a specific treatment. Tests such as CellSearch® and MammaPrint® fall into this category.
- Approve registrational endpoints that can be evaluated with commercialized diagnostic tests, such as LDL or viral load.
From 1998 to 2016, the U.S. Food and Drug Administration (FDA) approved 167 oncology drugs. But only about 10% of the treatments were accompanied by companion diagnostics. Examples include the cobas® EGFR Mutation Test v2 for use with osimertinib, PD-L1 IHC 22C3 pharmDX with pembrolizumab and FoundationFocusTM CDxBRCA with rucaparib.
Successful tests show a large effect in clinical trials –that is, patients with the predictive biomarker respond much better to treatment than those without it. These diagnostics are present in early development, and they usually test a single analyte using established technology such as PCR or immunohistochemistry.
Key case studies: YERVOY® and erdafitinib
One example of a drug that has benefited from pharmacodynamic biomarker studies is YERVOY® (ipilimumab). Approved by the FDA for metastatic melanoma in 2011, this treatment is an antibody that blocks human cytotoxic T-lymphocyte associated protein 4 (CTLA4).
To determine the mechanism of action, researchers studied biomarkers in the tissues of patients before and after YERVOY treatment. Staining the tumors with hematoxylin and eosin (H&E), showed that lymphocyte infiltration increased after the drug was administered. In patients who benefited from treatment, expression of the genes FoxP3 and IDO was higher at baseline than those that did not improve with treatment. The result with the H&E staining suggests that the drug draws more lymphocytes to the cancer; importantly, the research also demonstrated to the FDA that tumors treated with an immuno-oncology compound may grow before shrinking. The tests with IDO and FoxP3 indicated that the tumors that had previously been exposed to attack by the immune systems, were the ones deriving more benefit.
Predictive biomarkers are difficult to find because sponsors must simultaneously develop an effective drug and identify the correct biomarker. If developers do not know the predictive biomarker by Phase I, it is likely too late to incorporate such a test.
The drug erdafitinib, a small molecule FGFR inhibitor, is a relevant example. To find predictive biomarkers, about 240 cancer cell lines were treated to determine which tumors were sensitive or resistant to the drug. Overexpression of FGFR1, FGFR2 and FGFR4 was linked to sensitivity, while RAS/RAF mutations were associated with resistance. This was further validated in the Phase I clinical trial were patients who responded very well to treatment had mutations that caused overexpression of the FGFR pathway.
Aligning drug and diagnostic development
One key challenge is coordinating development of the drug and companion diagnostic so that they can be released at the same time. Pharmaceutical companies often rely heavily not only on a diagnostic partner but a CRO to manage the clinical research and assay development. Regulatory paths for the two products must be aligned, even though separate FDA groups handle their approval. Finally, partners need to perform proper validation of the assay at each development stage.
To define the right test, sponsors should consider three types of factors:
- Developers must identify the analyte to be assessed, the associated technology and the assay’s analytical properties such as specificity and sensitivity.
- This process spans sample collection to delivery of results. Key considerations include sample requirements and pre-analytic, post-analytic and time in motion steps. Even the difference between a serum and plasma sample can be crucial, and processing can sometimes add hours to the overall turnaround time.
- Sponsors need to assess the availability of technology in core markets, overall cost structure, adoption of equipment and reimbursement. For instance, if the test is too expensive or requires a cumbersome machine, it is unlikely to succeed.
Covance offers comprehensive biomarker services, ranging from biomarker discovery to commercialization of companion diagnostics. Our experts can perform more than 550 types of assays and provide access to solutions such as genomic testing, next-generation sequencing, anatomic pathology, immunohistochemistry and flow cytometry.