Q2 Solutions Q2 Solutions Actionable Insights for Better Health and Lab Solutions
Blog
Patrick Hurban, Ph.D. | January 23, 2017

The End to End Importance of Selecting the Right Biomarker in Cancer Studies

A critical component in the fight against cancer and optimizing treatment outcomes for patients
Researchers, healthcare providers, patients, families and friends join together on World Cancer Day to reaffirm our commitment to a unified fight against a disease that affects billions of people across the globe. Whether it be the patient who joins a clinical study, the friend or coworker that makes a charitable donation, the medical student who decides to specialize in oncology, or the company that commits to invest in the development of a new therapy, we all contribute how and where we can, according to our abilities and resources.

At Q2 Solutions, we are helping develop and use biomarkers in oncology trials across the development spectrum.  Cancer trials increasingly rely on biomarkers to optimize treatment outcomes for patients.  In fact, 60% of the 34 oncology drugs expected to launch in 2017 will be targeted to biomarker-defined patients.1 The importance of selecting the right biomarker early in the process of development cannot be overstated. Approaches involving the use of biomarkers must fit into the development continuum, starting at the preclinical stage and moving through to diagnostic development. 

Leveraging the right technologies and tools 
Q2 Solutions has unique genomic capabilities to optimize this process, including next generation RNA-Sequencing (RNA-Seq) for broad screening, which overcomes some of the limitations of previous tools for analyzing gene expression profiling of complex diseases.  Later in development, a more targeted approach can be used, focusing on a smaller group of high-value biomarker candidates that can be taken forward into quantitative polymerase chain reaction (qPCR) or targeted sequencing assays, either for expression or for identification and characterization of the variants in targeted genes. 

Q2 Solutions recently supported a study with a focus on ovarian cancer time-to-recurrence after adjuvant treatment of surgery followed by platinum-based chemotherapy. The initial feasibility study profiled 45 primary, recurrent and second-recurrent tumors using RNA-Seq, resulting in 22 RNAs of interest, associated with potential subtypes and time-to-recurrence. These were then used in a transition study, where the method was transitioned to a qPCR analysis, and used a naive cohort of 110 total tumors (71 primary, 39 recurrent). This identified two genes, which when expressed in a certain way, indicated that there was an enhanced probability of a longer time to recurrence.3

Recommendations for avoiding pitfalls and choosing the right patient cohort in cancer studies
For all the best intentions, there are many places along the way where a lack of careful planning and focus could result in substantial disturbance to timelines or outcomes.  Some common pitfalls to avoid in studies aiming at discovery of potential genomic biomarkers include:

  • Inadequate understanding of the underlying biology of the system involved
  • Having too few samples or ill-controlled confounding factors and not allowing for appropriate statistical power
  • Not using the proper study cohorts to allow for identification of biomarkers with the right discriminating power
  • Relying on poor governance of sample collection, handling and processing
  • Erroneous data analysis and interpretation (e.g., overfitting) 
  • Failing to identify early on how the potential biomarkers will be validated. 
The use of biomarkers in cancer studies requires a vigilant eye on the goals of each stage of the development continuum.

Additionally, it is vital to ensure that the cohort in the study is a good match for the intended use population, to minimize rates of false positives or negatives. Key questions to direct better outcomes include: 
  • Are the number of subjects adequate for the anticipated effect size? 
  • How robust and stable is the biomarker? 
  • How consistently and accurately can it be collected, preserved and measured? 
The likelihood of finding a true effect depends on the interplay between sample size, effect size, aggregate variability (biology, collection, analytical method), and type I and type II error probabilities. It is an incredibly exciting time to be involved in the development of cancer therapies. Biomarkers play a critical role in every phase of today’s cancer studies. With them, we can increase the probability of delivering life-saving cancer therapies to the patients who are most likely to benefit from them. On World Cancer Day, please join us as we contribute our time, talents, passion, and dedication to this important fight. 

For more information, see Best Practices for Integrating Biomarkers across the Drug Development Continuum


References: 
1McKesson Specialty Health/US Oncology Network (MSH) http://www.nxtbook.com/nxtbooks/pharmcomm/ 20130910/  

2Drucker E, Krapfenbauer K. Pitfalls and limitations in translation from biomarker discovery to clinical utility in predictive and personalized medicine. The EPMA Journal. 2013;4:7; http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3599714/

3Ganapathi M, Jones WD, et al. Expression profile of COL2A1 and the pseudogene SLC6A10P predicts tumor recurrence in high-grade serous ovarian cancer (2015). Int J Cancer. 2016 Feb 1;138(3):679-88. doi: 10.1002/ijc.29815. Epub 2015 Sep 10; http://www.ncbi.nlm.nih.gov/pubmed/26311224