Personalizing Cancer Drugs: The Next Front in Diagnostics

Patricia Lorusso, DO

Less than a decade after the FDA set the ground rules for developing assays that pair molecular targets with new drugs, experts say there have been strides in personalizing anticancer therapies but that many hurdles remain before next-generation sequencing and other precision medicine advances are incorporated into the diagnostic paradigm.

As it stands now, FDA-approved companion diagnostics that stratify patients for particular therapies based on a molecular marker have been the path to targeted therapy drug approvals and a mechanism that helps ensure reimbursement from payers.

Amid a competitive drug development market and technological advances, however, experts throughout the oncology field—including FDA officials—are wondering aloud about how to take the next step from the “one drug, one test” model and embrace emerging diagnostics that offer more information about tumors.

“Today, it seems as though ‘per drug’ testing is rapidly becoming impractical, as multiple development efforts converge on the same targets, and multiple targets are recognized within a disease entity that was previously not easily subclassified on a molecular basis,” according to Elizabeth A. Mansfield, PhD, who is the deputy director of personalized medicine in the Office of In Vitro Diagnostics and Radiological Health, which is part of the FDA’s Center for Devices and Radiological Health (CDRH).¹

Mansfield noted that multiplex testing, which provides information on more than one molecular marker, and the use of gene expression signatures as biomarkers may help solve the immediate problem of having to conduct multiple genetic tests on the same patient sample. She said the FDA would be flexible in evaluating new approaches.

Looking forward, Mansfield cited the development of next-generation sequencing technologies, which she defines as high-throughput DNA or nucleic acid sequencing, as particularly promising. However, she said the detection of rare mutations of unknown significance through sequencing remains an unresolved scientific question that would complicate the use of such diagnostics.

Even as the recently announced US government initiative and innovative precision medicine oncology clinical trials propel personalized medicine in cancer care forward, leading researchers and pharmaceutical industry observers expect a long road ahead in diagnostics.

The Current Landscape

“I do believe personalized medicine is still in its infancy,” Patricia M. LoRusso, DO, a professor of medicine and associate director of Innovative Medicine at Yale Cancer Center, said in an interview with OncologyLive.As a therapeutic class, oncology leads the way in the development of personalized medicine. Approximately 36% of the nearly 140 approved products with pharmacogenomic information on the labels are for oncology or hematology medicines, according to the FDA.² These biomarkers include germline or somatic gene variants, protein biomarkers used to select patients for therapies, and chromosomal abnormalities.

There also are several gene signature tests and sequencing assays that have been incorporated into treatment guidelines in evaluating patients for breast and ovarian cancers.

When it comes to pairing a particular drug with molecularly targeted therapies, however, there are specifically approved in vitro companion diagnostics, which the FDA defines as a “device that provides information that is essential for the safe and effective use of a corresponding therapeutic product.”³

Currently, the FDA has approved companion diagnostics for 15 anticancer drugs and one supportive care agent sometimes used for patients with myelodysplastic syndromes (Table). Those numbers are likely to increase markedly in the coming years, according to Joshua P. Cohen, PhD, a research associate professor at the Tufts Center for the Study of Drug Development in Boston.

Cohen said several dozen companion diagnostics are being developed in phase I and phase II oncology clinical trials, and that he expects 50% to 60% of newly approved cancer drugs to have a companion diagnostic within the next five years.

“There’s plenty of failure in any drug development process, so you’re not going to hear about those several dozen diagnostics nor the therapeutics that are linked to them,” Cohen said in an interview. “You’ll only hear about a few of them that make it through the process.”

Although progress in personalized medicine has not unfolded at the rapid pace once forecast when the human genome was sequenced in 2001, Cohen noted that the biopharmaceutical industry has significantly expanded its spending since 2010.

Investment in personalized medicine has nearly doubled during the past five years and the average company expects to increase its spending by 33% during the next five years, executives from leading biopharmaceutical businesses told Tufts researchers (without disclosing specific dollar figures).⁴

Challenges in Moving Forward

“Companies involved in personalized medicine report that across all phases of development, 73% of oncology compounds rely on biomarker data,” and more than 50 cancer drugs currently in latestage development are considered personalized medicines, the Tufts report indicates.Despite the increased investment in cancer diagnostics, a transition from the current “one drug, one test” model to technologies such as next-generation sequencing presents scientific and developmental challenges, Cohen noted. “If we know more specifically about the linkage between biomarkers A, B, and C, disease pattern Y, and patient X, then we can do the proper personalization and we can have a next-generation sequencing doing that for us,” said Cohen. “Until that time, I think we’re going to be dealing with one-to-one tests, however inefficient they appear.”

FDA-Approved Companion Diagnostics

^a Immunocytochemical HER2 assay also approved for eavluating patients with gastric cancer.

CISH, chromogenic in situ hybridization; FISH, fluorescence in situ hybridization; GIST, gastrointestinal stromal tumor; IHC, immunohistochemistry; ISH, in situ hybridization; NSCLC, non—small cell lung cancer; PCR, polymerase chain reaction.

Source: US Food and Drug Administration. List of cleared or approved companion diagnostic devices (in vitro and imaging tools). http://goo. gl/41ztzw. Updated October 7, 2015. Accessed October 13, 2015.

Cohen anticipates an increasing number of diagnostics co-developed along with a drug, as has been the case with ALK testing for patients who are candidates for crizotinib (Xalkori), and continuing identification of assays for previously approved drugs, such as the EGFR assay approved in 2014 for the 10-year-old cetuximab (Erbitux).

From her vantage point, LoRusso sees a need for next-generation sequencing in clinical trials, such as the Stand Up 2 Cancer Dream Team study in melanoma that she is leading, but not yet in clinical practice. The melanoma researchers are looking for mutations other than BRAF that can be targeted.

“The question as to whether or not next-generation sequencing will be replacing companion diagnostics is a very complex question,” said Lo- Russo. “Next-generation sequencing, especially the complex sequencing that we’re doing in our trial, is still quite investigational. Companion diagnostics do serve a purpose when there is a known drug and a known druggable target.”

The difficulty in translating next-generation sequencing into practice was illustrated recently in the phase II SHIVA study, LoRusso noted. The study, conducted at eight French academic centers, assessed molecularly targeted therapies outside of their label indications based on patients’ tumor profiles.⁵

Adult patients with metastatic solid tumors refractory to standard therapies were enrolled in the study if their tumors harbored molecular alterations in one of three pathways: hormone receptor, PI3K/AKT/mTOR, or RAF/MEK. After screening, 195 patients were randomly assigned to one of 10 molecularly targeted treatment options that matched their tumor profile or to treatment of physician’s choice.

After a median follow-up of 11.3 months, investigators found that the median progression- free survival was similar in the experimental and control groups (2.3 months vs 2.0 months, respectively) but the number of patients who experienced grade 3-4 toxicities was greater for those who received targeted agents compared with standard therapies (43% vs 35%, respectively). They concluded that “off-label use of molecularly agents should be discouraged.”

Making the Leap to Clinical Practice

“Although next-generation sequencing is becoming more widely available and accessible, as a tool to use for our patients in treatment decision making, except in rare situations the data obtained from that should not be acted upon outside of a clinical trial setting,” said LoRusso.For diagnostic companies seeking to translate next-generation technology into clinical practice, the challenges are both biological and regulatory. “Companion diagnostic testing—diagnostic testing in general—is certainly getting more complex as we know more about the biology,” Richard Watts, an executive at Qiagen, the global testing powerhouse, said in an interview with OncologyLive. “We’re seeing a lot of research scientists making use of the next-generation approaches to patient profiling.

However, from a regulatory perspective, we still face an upward battle or challenge.

“The challenge is that we have to validate our diagnostic assays by using clinical specimens which represent the content of these diagnostic tests, and FDA would like to see clinical specimens harboring each of the different aberrations that we could be looking for,” said Watts, who is vice president for Global Business Development, Companion Diagnostics, Americas, for Qiagen. “That makes it extremely challenging because you’re looking for a needle in a haystack, and it’s very difficult to find clinical specimens with very rare mutations,” Watts added. “So, when you’re talking about multigene tests for very rare events, the biology can be a bit of a roadblock.”

Nevertheless, Watts is confident that complex assays will make the transition from academic centers and research laboratories into clinical practice. “It’s not a question of will this effect a change and will this be implemented, it’s a question of when—and I don’t think we’re far off.”

Qiagen, for example, is a major provider of oncology companion diagnostics that also is breaking new ground in advancing next-generation testing.

The company’s portfolio currently used in clinical practice includes the therascreen line of companion diagnostics, notably single-gene assays for analyzing tumor tissue for EGFR and KRAS mutations. These tests were developed in collaboration with the pharmaceutical companies that make the corresponding drugs.

Lung Cancer Research Leads the Way

Qiagen also provides next-generation sequencing assays and bioinformatics analytics for researchers. Looking forward, Qiagen is developing its proprietary GeneReader device as a complete system that will be able to prepare samples for analysis, enrich for the most relevant target genes, sequence the DNA, and interpret the raw data.When it comes to charting the future of companion diagnostics in oncology, non—small cell lung cancer (NSCLC) is leading the field, thanks partly to a decade of advancements in understanding the molecular drivers of the disease. Since the discovery of the cancer-causing role of EGFR mutations in 2004, driver alterations have been identified in a dozen actionable genes in adenocarcinomas and about four genes in squamous cell carcinomas.⁶

As the menu of potentially actionable molecular targets has grown, the variety of testing options has grown beyond single-gene assays to multiplex testing that can interrogate a gene in greater detail. The cobas EGFR Mutation Test approved as a companion diagnostic for EGFR-targeting therapies for NSCLC is able to identify 41 mutations in EGFR across four exons.⁶

Given the difficulty in obtaining adequate tissue for analysis from patients with lung cancer, researchers have been particularly focused on developing panels that test for more than one gene at a time and on serum- or plasma-based “liquid biopsies.”

For example, the FDA is evaluating rociletinib (CO-1686) for patients with NSCLC whose tumors harbor mutations in EGFR T790M with Qiagen’s therascreen EGFR kit as a companion diagnostic. Additionally, plasma-genotyping using Sysmex’s BEAMing system is being explored as a diagnostic.

Watts believes multigene testing panels in lung cancer will become standard of care. “I just came back from World Lung in Denver and listened to a lot of clinicians talking about their workup in the diagnostic setting,” said Watts, referring to the 2015 World Lung Cancer Conference in September.

“They’re looking at either testing for EGFR and then reflexing to other actionable mutations in a panel or profiling the patient and then doing a confirmatory test to ensure that what they found in a panel is actually the clinical truth of that sample.”

Besides the new technology being developed, leading lung cancer researchers are calling for changes in way that molecular assays are standardized and validated. In a commentary in Clinical Cancer Research, Fred R. Hirsch, MD, and colleagues suggested the current system of companion diagnostics for new drugs is hampering research into assays particularly at a time when multiple companies are working on drugs for the same target.⁷

A prominent example of the industry competition and its impact on companion diagnostics is the race to develop immunotherapy agents that inhibit the PD-1/PD-L1 pathway. Four major companies are working on NSCLC therapies using different assays to measure PD-L1 expression in the drugs they are testing, prompting the FDA and major oncology associations to bring together the drug developers, diagnostic companies, research leaders, and advocates for an unusual workshop to discuss how best to move forward with new diagnostics in the molecular era.

One of these agents, pembrolizumab (Keytruda), was approved by the FDA in early October for the treatment of patients with NSCLC whose tumors express PD-L1 as measured by a PD-L1 immunohistochemistry test from Dako. Meanwhile, industry efforts to harmonize PD-L1 assays continue. Hirsch et al also noted the need to streamline the path to development of multiplex assays including next-generation sequencing techniques.

Roy S. Herbst, MD, PhD, who has helped pioneer molecularly targeted research in NSCLC and was a coauthor on the paper, said in an interview that he believes such advances eventually will be incorporated into clinical practice.

“We do need to multiplex our tests and tissue is scarce,” said Herbst, who is chief of medical oncology at Yale Cancer Center and Smilow Cancer Hospital at Yale-New Haven. “One test— one drug is a nice aspiration, but I think we’re probably going to want to multiplex, do many tests at one time, and then use that to pick the drug or a combination.”

Change, Herbst said, “is around the corner— but it’s a long road.”

References

1. Mansfield EA. FDA perspective on companion diagnostics: an evolving paradigm. Clin Cancer Res. 2014;20(6):1453-1457.
2. US Food and Drug Administration. Table of pharmacogenomics biomarkers in drug labeling. http://goo.gl/d0jHb. Updated May 20, 2015. Accessed October 6, 2015.
3. US Food and Drug Administration. In vitro companion diagnostic devices: guidance for industry ad Food and Drug Administration staff. http://goo.gl/wCJ6oZ. Published August 6, 2014. Accessed October 5, 2015.
4. Tufts Center for the Study of Drug Development. Personalized medicine gains traction but still faces multiple challenges. Impact Report. May/June 2015.
5. LeTourneau C, Delord JP, Goncalves A, et al. Molecularly targeted therapy based on tumour molecular profiling versus conventional therapy for advanced cancer (SHIVA): a multicentre, open-label, proof-of-concept, randomised, controlled phase 2 trial [published online September 3, 2015]. Lancet Oncol. 2015;16(13):1324-1334.
6. Naidoo J, Drilon A. Molecular diagnostic testing in non-small cell lung cancer. Am J Hem Oncol. 2014;10(4): 4-11. www.gotoper.com/publications/ajho/2014/2014Sep/Molecular-Diagnostic-Testing-in-Non-Small-Cell-Lung-Cancer.
7. Hirsch FR, Bunn PA, Herbst RS. “Companion diagnostics”: has their time come and gone? [published online July 24, 2014]. Clin Cancer Res. 2014;20(17):4422-4424.