MD+DI Online is part of the Informa Markets Division of Informa PLC

This site is operated by a business or businesses owned by Informa PLC and all copyright resides with them. Informa PLC's registered office is 5 Howick Place, London SW1P 1WG. Registered in England and Wales. Number 8860726.

AI Mines Clinical Data for Thyroid Cancer Therapeutic Targets

AI Mines Clinical Data for Thyroid Cancer Therapeutic Targets
Image by Jae Rue from Pixabay 
An artificial intelligence powerhouse and a leader in molecular diagnostics have teamed up to increase the precision of thyroid cancer diagnosis and minimize unnecessary treatment.

The Helomics division of Predictive Oncology and Interpace Diagnostics Group Inc. have announced a cooperative research agreement. Helomics has developed an AI bioinformatic platform, known as “D-CHIP,” while Interpace has collected an impressive cache of clinical data from next-generation sequencing (NGS) and microRNA testing of thyroid tumor samples from its combination ThyGeNEXT panel and ThyroMIR classifier. By analyzing the Interpace datasets on the D-CHIP platform, Helomics and Interpace plan to improve diagnosis of thyroid cancer and identify druggable targets.

The ThyGeNEXT and ThyraMIR combination testing platform for genetic alterations and micro RNA expression detects markers for specific types of thyroid cancer. Traditionally, thyroid cancer was diagnosed by cytology reads from fine needle aspiration (FNA) biopsy of suspicious thyroid nodules. However, up to 30% of FNA samples are cytologically indeterminate. Furthermore, thyroid tumors range from relatively benign and slow growing to aggressive tumors with a high likelihood of metastasis. Interpace has selected markers for its panels that are specific for individual subtypes of thyroid cancer.

“We want to choose markers that matter,” Alidad Mireskandari, PhD, vice president of business development of Interpace, told MD+DI. “TERT and ALK,” he pointed out, “are markers of aggressiveness.” These markers are hallmarks of large tumors with a high likelihood of metastasis (especially to the lymph nodes), post-treatment recurrence, and death.

“We are the only testing company with two platforms,” Mireskandari clarified, “and we have validated multiple specimen types, including FNA, cytology slides, and formalin-fixed paraffin-embedded samples.” The ThyraMIR panel, which measures changes in expression of 10 microRNAs, is designed to work in concert with ThyGeNEXT. MicroRNAs are small RNA molecules—only 21 to 23 nucleotides—that are noncoding but serve as master regulators, governing the expression of complex networks of proteins.

The results allow physicians to accurately classify benign and malignant nodules and thereby improve preoperative risk-based management. “In clinical trials,” Mireskandari stressed, “the combination testing reduced unnecessary surgeries by 85%.”

Interpace’s molecular diagnostic panels are continually evolving as new and more predictive markers are discovered. The partnership with Helomics will turbocharge this scientific discovery process, winnowing through gigabytes of data to tease out significant associations.

“We have four to five years’ worth of data from more than 10,000 samples sequenced and profiled,” Mireskandari explained. The Helomics model is expected to not only yield additional markers but also map pathways of pathogenesis and disease progression. Proteins along these pathways may prove to be new therapeutic targets.

“My hope is the Helomics model will allow us to further optimize performance of our tests and identify druggable targets. We are not a therapeutics company,” Mireskandari clarified. “We will partner with a third party for development of any new therapies.” Interpace has, however, just acquired the Biopharma Services business of Cancer Genetics Inc., expanding its services to oncology and immuno-oncology drug development customers.

TAGS: News
Hide comments


  • Allowed HTML tags: <em> <strong> <blockquote> <br> <p>

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.