The plan to make cancer predictable
Researchers with the APOLLO platform of the Moon Shots Program are using immense amounts of longitudinal clinical and research data to understand why the disease responds to or resists certain treatments
When the ever-growing arsenal of new cancer therapies meets the ever-changing target of advanced cancer, the outcome often is unpredictable.
Immunotherapies, which free the immune system to attack tumors, result in long-term remissions, even cures, for a significant but small number of patients across several cancer types. But for most, they have little or no impact.
Right now, there’s no reliable way to predict who will benefit from immunotherapy treatment. Or to envisage what the next best step should be after a treatment of any type — immunotherapy, targeted therapy, chemotherapy, radiation — fails.
Getting a firm scientific grip on the extreme survival skills of advanced cancer is the mission of MD Anderson’s APOLLO program, which stands for Adaptive Patient-Oriented Longitudinal Learning and Optimization.
“Cancers are exceptionally adaptive in the face of treatment pressure,” says Andrew Futreal, Ph.D., interim chair of Genomic Medicine and co-leader of APOLLO. “We know a tumor after therapy is likely to be different than before, but we haven’t done a good job of tackling this from a research perspective until now,” says Futreal, who also co-leads MD Anderson’s Moon Shots Program.
APOLLO is a systematic approach that relies on taking blood samples and biopsies before, during and after treatment, and conducting deep molecular and immune analysis of those tumor samples to understand what causes them to respond to or resist a given treatment.
In the next two years, APOLLO is scheduled to conduct such analyses in 2,100 patients enrolled in 28 high-priority clinical trials for melanoma, multiple myeloma, glioblastoma, lymphoma, sarcoma, lung, breast, colorectal, pancreas and
ovarian cancers, and cancers caused by the human papillomavirus.
GET TO KNOW APOLLO
APOLLO stands for Adaptive Patient-Oriented Longitudinal Learning and Optimization. It’s a platform for MD Anderson’s Moon Shots Program, which was created to save more lives by accelerating the development of new approaches based on scientific discoveries.
APOLLO addresses advanced cancers’ tenacious ability to survive by changing in response to treatment by systematically:
- Gathering high-quality biopsies and blood samples before, during and after treatment.
- Preparing samples for deep genomic and immune response analysis by expert platforms.
- Depositing analytical results, with clinical information and other data, in the Translational Research Accelerator (TRA), a comprehensive and secure big data platform.
MD Anderson researchers will be able to search the TRA to develop new hypotheses and science-based therapies.
By the way: In Greek mythology, Apollo was the god of healing and medicine, light and truth.
Beyond the initial biopsy
Research with serial biopsies is important to understanding what is happening with cancer on a molecular level and building a knowledge base leading to the routine use of repeat biopsies in the clinic.
“If you don’t get a biopsy after treatment, you’re not going to learn about tumor evolution and how tumors resist treatment,” says Ignacio Wistuba, M.D., APOLLO co-leader and chair of Translational Molecular Pathology.
Cancer patients generally undergo an initial biopsy to guide treatment, with shrinkage or progression measured afterward by CT or MRI scans for size and volume, and PET scans for metabolic activity.
Biopsies may be taken surgically, through endoscopy, or with radiologically guided core needles or fine needles that extract small samples from less accessible tumors and, potentially, from circulating blood. Interventional radiologists take image-guided core-needle biopsies in difficult-to-access tumors.
“MD Anderson’s interventional radiology group is one of the best in the nation and they’ll play a key role in APOLLO,” Wistuba says.
While taking biopsies after treatment makes scientific sense, there are risks with the procedures, and the clinical benefit of doing so hasn’t been proven, which means insurers and other payers don’t currently cover the cost.
“We have to build the evidence first,” Futreal says. “You can’t just go full bore into clinical implementation of this until we generate and analyze the data in the research setting.”
Patients enrolled in clinical trials under the APOLLO protocol are asked to consent to repeat biopsies and multiple blood draws at no additional cost. They may also refuse to have biopsies at any time during the study.
The development of liquid biopsies — blood tests that allow lessinvasive analysis of tumors before and after treatment — will be a perfect fit for this approach, Futreal notes.
Gathering all the data
Results of these analyses, plus clinical information, will be added to the Translational Research Accelerator (TRA), which is a big data platform that integrates longitudinal clinical and research data — the same sample tracked at different points in time — to support translational research throughout the institution.
Information from approximately 250,000 patients treated at MD Anderson since 2012 has been loaded into the secure database. Research data from Moon Shots Platforms is being added, with the full platform being available in fall 2016.
The TRA will provide MD Anderson researchers with an unprecedented capacity to more quickly and efficiently generate science-based inquiries in the pursuit of better cancer treatment.
“APOLLO, with its pipeline of comprehensive data feeding the Translational Research Accelerator, drives our vision of every patient contributing to, and potentially benefiting from, research,” says Futreal.
APOLLO pilot: Biopsies during treatment tell new tale for melanoma
Immune response measured in tumor biopsies during the course of early treatment predicts which melanoma patients will benefit from specific immune checkpoint blockade drugs – drugs that release a molecular brake on the immune system, freeing it to fight cancer – MD Anderson researchers reported in the journal Cancer Discovery.
Analysis of biopsies before treatment did not indicate who would respond in this unique longitudinal study of 53
melanoma patients treated with two immune checkpoint inhibitors between October 2011 and March 2015.
“Before treatment, analyzing samples with a 12-marker immune panel or a 795-gene expression panel, you can’t
tell who will respond with any degree of certainty. During treatment, there were night-and-day differences between
responders and nonresponders,” says study senior author Jennifer Wargo, M.D., associate professor of Genomic
Medicine and Surgical Oncology. She is also the co-leader of the clinical aspects of APOLLO.
Response rates ranged from 8 to 44% of patients given the drugs separately, with many responders having completeresponses that last for years. Identifying biomarkers to help determine who should receive these drugs has been thesubject of much research, but biomarkers have not strongly or exclusively predicted response before treatment.
The team’s findings suggest that assessment of immune responses should be considered in biopsies taken shortlyafter treatment begins, Wargo says. This is because they may provide far more value than analysis of retreatment samples, at least until better pretreatment biomarkers are identified.
Their findings could also help guide treatment with drugs that block protein receptors on T cells known as PD-1 and CTLA-4. Both of these receptors shut down the immune system’s attack on tumors.
Serial biopsies identify signature of success
“Profound and highly statistically significant” differences between responders and nonresponders to anti-PD1 therapy were found in nearly all of the 12 immune markers in the early on-treatment biopsies. These included the density in the tumor of killer T cells, other T cells that assist the killers, and the presence of immune checkpoint molecules.
The team’s findings have implications for treatment and further research to understand how melanoma responds to or resists treatment.
“We could start by treating with anti-PD1, do an early on-treatment biopsy and, based on that, either continue or add another agent,” Wargo says.
Tumor samples were collected at various times when it was technically feasible and safe to do so. For example, of the 46 patients undergoing anti-PD1 treatment, 24 had pretreatment biopsies (seven responders, 17 nonresponders), 11 had biopsies during treatment (five responders, six nonresponders) and 12 provided tumor samples at progression.
“We should be incorporating analysis of longitudinal tumor and blood samples into clinical trials, and, ultimately, we may even incorporate this into treatment with standard-of-care therapy,” Wargo says. “This effort is critical to guiding patient care in this era of precision medicine.”