One Friday evening in 2011, just before Halloween, Dr. Byron Hann was alone in his lab at the University of California San Francisco (UCSF) when he got a call from UCSF’s Brain Tumor SPORE Biorepository, part of the Cancer Center’s Neurologic Oncology Program. Aware that Dr. Hann had been attempting to create a patient derived xenograft (PDX) model for chordoma – human tumor tissue grown in specialized mice – they asked if he’d be interested in excess tumor tissue that was being removed from a chordoma patient in the operating room.
Dr. Hann, whose lab focuses on creating PDX models to test new cancer therapies, emphatically accepted. As the only one left in the lab that night, he raced to the operating room to retrieve the sample, and then back to his lab, where he implanted it in specialized mice. Then he waited. And waited.
A few months later, the tumor he had implanted in the mouse started growing. He harvested the tumor from one mouse and implanted it in several more. Months later these tumors grew too. He sent tumor samples from these mice to Dr. Joanna Phillips, the neuropathologist at UCF’s Brain Tumor Biorepository, and sure enough, they looked exactly like a human chordoma under the microscope. Molecular tests confirmed that the biology of the tumors in the mice matched human chordoma too. After propagating the tumors once more into a third set of mice, the tumor continued to grow and maintain the appearance of a human chordoma. They had succeeded in creating what would become one of the first chordoma PDX models ever developed. Its name: SF8894.
“It was very exciting to realize we had a viable model,” recalls Dr. Hann. “Once we confirmed that it was really chordoma, I called Josh [Sommer], whose story had inspired me to focus on chordoma in the first place, to plan our next steps.”
After independently validating the new PDX model, the Foundation awarded a $10,000 prize to Dr. Hann and his team (The UCSF Preclinical Therapeutics Core) in recognition of their important contribution to chordoma research. Since then, Drs. Hann and Phillips have gone on to publish their results in the Journal of Neurosurgery, implant several more tissue samples, and recently generate a second PDX model, which was validated in 2016.
Critical tools for testing the potential of new therapies, these new PDX models are now being used to help identify the most promising drugs to move into human clinical trials.
“The whole process has involved people who have been very unselfish about ownership and authorship,” said Dr. Hann. “It’s wonderful to be part of a group that sees the big picture – improving patient care.”
Ramping up model development
While a combination of prizes and persuasion had spurred several researchers like Dr. Hann to attempt to develop new PDX models, the success rate remained low. By 2014 the Foundation recognized that more attempts were needed to rapidly develop a robust panel of chordoma PDX models for drug testing.
To meet this need, the Foundation partnered with South Texas Accelerated Research Therapeutics (START), a Texas-based contract research laboratory with a track record of establishing difficult to grow PDX models. Through this partnership, when patients contribute their tumor tissue through the Foundation’s Tumor Donation Program, a portion of the tumor is sent to START to attempt to establish a PDX model.
What makes this effort particularly unique is the fact that it is set up to collect tumor tissue directly from patients. This means that any models resulting from the implantation of patient-donated tissue belong to the Foundation, and can, therefore, be shared directly with researchers anywhere in the world without the lengthy legal negotiations typically involved in transferring materials between universities.
A streamlined process
The PDX development process is overseen by Dr. Michael Wick, a molecular pharmacologist who founded and directs the preclinical research division at START. Dr. Wick and his team work with the Foundation’s manager of research, Patty Cogswell, to coordinate chordoma tissue collection at the surgical site, providing special kits designed to store viable fragments of tumor tissue and keep them alive during transport to San Antonio.
Once the tissue arrives at START, it is implanted into specialized mice and then carefully monitored for growth. If the tumor grows to a usable size, it is removed and tested to confirm that the resulting tumor is, indeed, a chordoma and is similar to the patient’s tumor from which it was derived. Pieces of the tumor are then implanted and grown in more mice, and again removed and validated. After being passed into a third round of mice, it can be expanded for use in drug testing experiments. Samples are also banked for future use or for distribution to other researchers
“One of the most striking things for us about working with the Chordoma Foundation is how many tissue samples we’ve gotten relative to the total number of chordoma surgeries being performed,” said Dr. Wick. “When we first embarked upon this partnership we expected to get one sample every few months, but after just one year, we’re getting a few samples every month, which is amazing. The number of surgical partners is increasing, and the Foundation continues to do a wonderful job of educating patients about tissue donation and engaging surgical sites in this process.”
A growing repertoire of PDX models
The Foundation’s goal is to develop a panel of PDX models that are representative of all the clinical variation seen in chordoma patients, including tumors of various anatomic locations, treatment histories, and patient ages.
To date, START has engrafted over 60 tumors, already surpassing the Foundation’s goal of 50 by the end of 2017. Among these, four have resulted in viable PDX models, 31 have not succeeded, and 26 remain in development.
Three of these models, plus Dr. Hann’s SF8894 model, are now available to the research community through the Foundation’s PDX Repository. They are also now available for use through the Foundation’s Drug Screening Pipeline. Several additional models are being developed by labs across the world and will be added to the repository as soon as they are obtained and validated.
As a result of these efforts, within the next one to two years it is likely that scarcity of PDX models will no longer be an impediment to chordoma research, solidifying yet another critical piece of research-enabling infrastructure needed to level the playing field for chordoma drug development.