Prostate Cancer Lab #30: Using Functional Profiling to Predict Drug Response (Dr. Robert Nagourney)

Advanced cancer patients face complex treatment decisions. Screening drug options by putting the drugs on tumor tissue (functional testing) is a seemingly obvious way to gather insights on what might work and what might not. Drug combinations, dosing options, and random batches of FDA-approved drugs can be run through a functional testing process, often uncovering novel insights. Most clinicians are skeptical that there is any information to be gained from “ex vivo” (outside the body) tests. They argue that the only valid evidence comes from experimenting with patients. But advanced cancer patients are looking for any edge that can help guide them to a treatment that might be more likely to generate a durable response.

Robert Nagourney, MD, is uniquely qualified to provide insights on the value of functional testing. He is a rare artisan in the medical field: an experienced physician treating patients who also runs a lab to guide treatment decisions, The Nagourney Cancer Institute. He bridges from testing to treatment, from pathology to oncology. Over the last two decades he has pioneered functional testing as a means to predict which chemotherapies and drug combinations will be effective for patients. He helps patients who fit a profile, e.g., those who have advanced disease and easy access to fresh tumor tissue, and guides their very personalized care based on his extensive experience, using functional testing and recently metabolomics (analysis of how cells get energy).

How should we think about diagnosing and treating cancer?

Cancer is complicated. To help advanced cancer patients make complex decisions about which treatments will work, we have to listen very carefully to what the tumor is telling us. Most advanced cancer patients get genomic sequencing which identifies unique biomarkers for a patient. Sometimes there is no biomarker or an agent for that biomarker. More often, there are multiple biomarkers identified. To hit a biomarker target, you need confidence that the treatment will work. You can’t depend solely on the models and platforms that use genetic analysis to pick drugs. Cancer is not a gene, a mutation, an amplification, or a fusion. Cancer is a cell that has advantages that allow it to do what it’s doing to you. All it wants to do is stay alive, and we’re trying to kill it.

We must interrogate the disease at a functional level – does the therapy kill the cancer cells? We have a tendency to want to believe that cancer cells behave according to our predictive models. We need to be humble and add testing, particularly functional testing, to add confidence to the treatment hypotheses that come out of genetic, transcriptomic, and proteomic analyses and other sources.

Is therapy screening using functional testing worthwhile – are its predictions accurate? How do you convince skeptical patients and treating physicians that these tests are worth pursuing?

The Nagourney Cancer Institute/Rational Therapeutics has conducted a meta-analysis of clinical outcomes in peer-reviewed literature for over 2500 patients with solid tumors they have treated. It shows that their predictive therapy recommendations (based on their functional analysis) have provided higher reductions in the tumor burden (over 2x) and higher one-year survival (1.4x).

Who can get functional testing? What does functional testing need as input?

Functional testing needs fresh tissue for treatment analysis. There are several possible inputs to test drugs on cancer cells: cell lines, organoids, FFPE tissue, and fresh tissue. “Cell lines” are cells that have been extracted from humans and then manufactured in the lab to grow endlessly. They are used for research. “Organoids” are similarly harvested cancer cells that maintain more of the tumor microenvironment. They can be used to guide treatment decisions for individual patients. “FFPE” (Formalin-Fixed Paraffin-Embedded, a method of tissue preservation) is tumor biopsy tissue that has been preserved for future analysis. Almost every patient has a tissue biopsy when their cancer is diagnosed, and some of that tissue is preserved. “Fresh tissue” is an untreated tissue sample that attempts to maintain as much of the environment around the cancer cells as possible. The advantage of cell lines and organoids is that they can be manufactured in quantities as needed for testing. The advantage of FFPE is that it is available whenever needed for additional analysis. The advantage of fresh tissue (and disadvantage of cell lines, FFPE, and organoids) is that it most closely simulates the cancer’s natural state and environment in the patient. Ideally these are human, three-dimensional tumors that give the features of human cancer that would predict outcomes in solid tumors. The disadvantage of fresh tissue is that not all patients will have readily available cancer in soft tissue in adequate quantities for testing.

The other input to functional testing is treatment hypotheses, which can be derived from genomic, transcriptomic, and proteomic analysis, and other sources, including drug combinations and dosing variations, or even random panels of approved drugs, including off label uses of approved drugs.

How can patients access functional testing?

Functional testing requires a cubic centimeter of fresh frozen tissue. The fresh frozen tissue needs to be shipped by overnight courier from the clinic where the biopsy or surgery is done to the testing lab. The lab can also run the tissue through microenvironmental metabolomics and take a blood draw to analyze biochemistry. The costs may be reimbursed by insurance and may be free.

What’s next in cancer testing?

Cancer is a metabolic disease. Cancer cells, like all cells, are making and using energy. They require glucose for energy and cell structure, lipids for energy and membrane structure, and amino acids for energy and protein structure. Tissue and plasma can be analyzed to gauge metabolism through mass spectrometry. The results can be used for diagnosis, prognosis, and prediction of response, moving beyond what can be learned from genomics, transcriptomics, and proteomics.