How Do I Measure Thee...
Safety Assessment
Regina Kelder

How Do I Measure Thee...

Biomarkers are ubiquitous in research but they don't always tell you what you need to know. The third of a four-part series.

Biological markers or simply biomarkers refer to a broad subcategory of medical signs—from in vitro assays of tissue and blood to molecular imaging, lipid tests and genetic mutations—that can be measured accurately and reproducibly.

The field has expanded quantitatively and qualitatively on both the clinical and preclinical research sides. Biomarkers are being used to show who is at risk for cancer and who is at risk for suicide, and different biological signatures are being combined to construct disease models. The more novel the biomarker, the more difficult validation can become.

SOT2015 LogoThe promise and the challenges of this burgeoning field were on full display this week at SOT2015, where many sessions and countless posters, either directly or indirectly, pointed to the important role of biomarkers in advancing clinical efficacy and safety assessment studies, and the challenges that the field faces in identifying biomarkers that are actually tell you what you need to know.

This is particularly evident in the development of cancer immunotherapies, particularly checkpoint inhibitors such as Yervoy® and Opdivo®.

Yervoy® (ipilimumab) and Opdivo® nivolumab are monoclonal antibodies that help people mount a robust response to tumors by unleashing the body's natural brakes on the immune system. Yervoy inhibits a protein on T cells called cytotoxic T lymphocyte antigen-4 (CTLA-4) while nivolumab blocks the action of programmed death 1 (PD-1), which is found on T cells and B cells.

The drugs, which are referred to as checkpoint inhibitors, are both approved for the treatment of melanoma. Recent studies suggest when you combine the two mAbs, the antibody cocktail works even better, and not just in melanoma but renal cancer as well.

The problem is only a small percentage of patients will respond favorably to the drugs, and scientists still do not have a good biomarker predicting who they are, noted Willem Overwijk, a melanoma researcher at MD Anderson Cancer Center, during a SOT2015 session on cancer immunotherapy.

Measuring tumor expression in PD-L1, the ligand of PDL-1, would seem like a logical choice says Overwijk, were it not for the fact that some of the patients who had PD-L1+ positive tumor did not respond to the combo of ipilimumab/ nivolumab while some with PD-L1- negative tumors did.

A more powerful predictor of who might respond to these checkpoint inhibitors might be mutational load, says Overwijk. A recent study found that melanoma patients with high mutational loads responded much better to ipilimumab than those who did not, possibly because the more mutated tumors were not recognized by the immune system and therefore viewed as a threat. In fact, tumors with very high mutational rates—like melanoma, lung and renal cancers—seem to respond more favorably to immunotherapies compared to colon tumors, which don't mutate as much. "We see this in mice, too," says Overwijk. "These mutation proteins tend to be the real focus of immune responses."

The preclinical arena

Given the high percentage of experimental compounds that end up being shelved because of toxicity issues, identifying reliable biomarkers and selecting the right assays are particularly important in preclinical settings. Biomarkers can be used to validate targets, illustrate the mechanism of action, predict therapeutic efficacy, help select the right patients and alert potential toxicity, said Simon Lavallée, Scientific Director, Biomarkers at Charles River during his SOT2015 exhibitor's presentation.

Lavallée said biomarkers can take the form of soluble analytes, physiologic measurements, behavioral assessments or various imaging modalities. Many classes of biomarkers reflect the presence of a specific molecular target, others are a step removed and measure changes in the subcellular phenotype of the organism that lead to alterations in proteins. Biomarkers are also used in disease diagnosis or disease staging.

While there is no shortage of kits on the market, finding the most relevant or reliable ones to use in research studies can be a frustrating exercise. It's important to ask the right questions and partner with reputable companies, he said. Do you know which publications have used the kit and reagents, and whether the kit was used across several species and if the data is actual or anecdotal? Will the manufacturer of the kits provide a reference range with the kit/historical data and are there sufficient reagents in the kit to allow optimization and validation?

If a manufacturer is offering assays for markers in species that have been reported as not to exist, consider that suspect, he noted. Other questions worth asking when looking for reagents:

  • Are protocol booklets appearing to be copy pasted versions across all assays?
  • Are all information about kit components proprietary?
  • Do the list of available assays appear to good to be true?

Lavallée likens the revolution in biomarkers to the onslaught of "Apps" that fill up our SmartPhones. They may look and sound attractive, but how many are actually useful?