The Allina Health Approach to Genomic Testing for Cancer

Sara Kerr, MD

In the past, oncologists had limited treatment options, such as surgery, chemotherapy, and radiation therapy. Now, genomic testing has transformed cancer treatment by allowing for more precise diagnosis and targeted treatment.

Experts can use genomic testing to identify specific mutations in a patient's cancer cells and target those mutations with the best agent they see fit. Different types of genomic testing available include fluorescence in situ hybridization (FISH) testing, polymerase chain reaction (PCR)-based testing, and next-generation sequencing.

While genomic testing also has some limitations, like the fact that it only provides a snapshot of the cancer at the time of testing, the field is rapidly evolving.

“I think this is only going to keep growing, getting better, and more precise in terms of what we can offer patients for therapy,” explained Sara Kerr, MD, in an interview with Targeted Oncology^TM.

One specific program, the Allina Health next-generation sequencing program, is particularly unique as it uses small biopsy specimens, such as cytology smears, to get genomic information for patients, which allows for rapid testing and quick results. According to Kerr, a pathologist at Allina Health, the program also ensures equitable access to genomic testing for all patients across diverse communities.

In the interview with Targeted Oncology^TM, Kerr discussed the evolution of genomic testing for patients with cancer and highlighted the program used at Allina Health.

Targeted Oncology: In recent years, how has genomic testing transformed cancer treatment?

Kerr: I started my medical training in the early 2000s and as a pathologist, when we were diagnosing cancer, we made a diagnosis, and then the patient went on to either having surgery or chemotherapy or radiation therapy, and there was not a whole lot else. In that time period, there was a drug called imatinib [Gleevec] for blood cancer, and that was 1 of the first targeted treatments beyond what we had with surgery, chemotherapy, radiation therapy, and that started changing the landscape of choices for patients and in cancer treatment. After that, many other targeted treatments started coming out, and with that genomic data, not only did it change how the choices that patients had for cancer treatment, but it also changed the way pathologists think about cancers and classifying cancers. Over time and during my career, it has been integrated into the way we classify cancers, so the diagnosis, the genomics, and the treatment are increasingly melding together [to make] a more integrated approach.

Big genomic data visualization: © majcot - stock.adobe.com

What are the different types of genomic testing available?

That has changed over time too. We had very simple technologies in the early days, which basically consisted of stains that we looked at under a microscope, FISH testing, which also involved microscopy, and PCR-based testing for single genes is how we first started out. There [is] such a huge variety of technologies that we use to accomplish those PCR tests. Now that has been streamlined mostly into next-generation sequencing, so highly parallel sequencing. We still do, in addition to next-generation sequencing stains that we look at under the microscope for protein, so PD-L1 is 1 that we do a lot, estrogen receptor, progesterone receptor, HER2, and things like that we still look at under a microscope mismatch repair proteins. Everything else has now been streamlined into next-generation sequencing, which gives us a lot of information from a small amount of cancer cells.

What are some potential benefits and limitations of using genomic testing in cancer treatment?

There are a lot of benefits from using a genomic approach that helps us with the precise diagnosis of cancer. It helps us find potential targets for treatment, especially in patients with advanced cancer. Sometimes it helps us determine whether cancer is going to be more or less aggressive. Some of the limitations, though, are that we are only getting a small snapshot of the cancer. So, with a small biopsy, there might be a cancer that is widely spread, and we are just getting a few cells from that cancer, and the cancer might vary from a primary site to a metastasis. That is where technologies like liquid biopsy or cell-free DNA are coming in, because it might give us that better overall snapshot of the cancer.

The other limitation at this point is that sometimes we do not find anything that is targetable. That can be good and bad. It puts patients on a path potentially leading to standard chemotherapy and radiation therapy, which can work well for some patients, but there is always a chance that when we do testing, we might disqualify patients from targeted testing strategies. It is just something to be aware of, especially for patients who are looking for hope.

Can you provide an overview on the Allina Health next-generation sequencing program and what differentiates it from others?

I joined Allina Health in 2019, and I came from an academic background where we were on the forefront of next-generation sequencing development. When I came to Allina Health, the program was pretty small, and next-generation sequencing had not started here. I saw an opportunity for a community hospital to get involved in using the small biopsies that are collected every day within our system and getting that sort of baseline information for patients in terms of genomics.

We have developed a program here that uses those small biopsy specimens, especially cytology smears, which often are not used by some of the larger commercial assays, and we use those small specimens to get that baseline information for most patients who qualify for genomic testing to find a target. For example, lung cancer is a good example where patients who have metastatic lung cancer, we test the cytology smears that we make from endobronchial ultrasound-guided fine-needle aspiration or targeted biopsies that are done in radiology, and we use the cytology smears preferentially to do our next-generation sequencing, just to give that baseline of genomic information that can decide whether patients go down at targeted treated treatment pathway, standard chemotherapy, chemotherapy and radiation, or chemo-immunotherapy. We also do PD-L1, which is a stain, at the same time. We do that in a rapid fashion with our in-house lab. Rather than having to prepare material to send out and get information back, we can get that going right at the time that the pathologist is looking at the slides and making the diagnosis. We get that next-generation sequencing and PD-L1 going immediately, even sometimes before the pathology report is out, and those results are usually back within about a week. We try to time that so that the patient has that information in time for their appointment.

If more information is needed from a larger commercial test, we try to preserve that material that is leftover after we have made the diagnosis, and we use a very small amount of material for in-house testing. Then that material can be sent out for more information from a larger panel. Our program is getting that targeted baseline information that patients and oncologists need right away.

What types of cancers and patients are eligible for testing within the program?

Basically, every type of cancer is eligible, but there are certain cancers that benefit more from our technology vs maybe something more comprehensive. Common cancers like lung cancer have a lot of treatment targets; [for] colorectal cancer, we do a baseline small panel such as immunohistochemistry; for melanomas we test for BRAF upfront; for metastatic melanoma, gastrointestinal stromal tumors, we look in patients who are eligible for neoadjuvant therapy. Those cancers that have sort of some of the common targets that are where patients can be matched with FDA-approved therapy. We test those cancers according to strict guidelines where the pathologist can recognize okay, this is the diagnosis, this is the stage, I need to do this baseline testing for the patient. So those are the most common cancers.

Some cancers require a more comprehensive approach. For example, ovarian cancer has some targets including the homologous recombination repair pathway where it requires a more comprehensive approach to get the full picture and for those cancers. We recommend sending out for the more comprehensive testing rather than using our targeted approach that we can use for the things that are sort of the low hanging fruit, the easy things that we can do in-house. Then for some types of cancers that require a more comprehensive approach or maybe rare cancers, we usually recommend sending those out for the larger test.

How does the program ensure equitable access to genomic testing for patients across Allina Health's diverse communities?

That is something that I am proud of with our program. As I described, we do not look at a patient's financial status when we are deciding who gets the Allina targeted testing. Patients qualify based on their diagnosis, and maybe how far their cancer is spread. We are not qualifying who does and does not get testing other than that. That sort of inexpensive baseline testing is done for everybody who qualifies based on their diagnosis and stage upfront. Then, that information is available to patients and oncologists in a very equitable way across the system. I am proud that we have been able to accomplish that and in a financially viable way for Allina Health.

Can you give any specific examples of how genomic testing has guided treatment decisions in patients within the program?

The most common scenario that I see is lung cancer, so pulmonary and non-small cell carcinoma and patients who have basically anything except for early surgical resectable disease get our testing. For example, if a patient is identified to have an EGFR mutation and has stage IV lung cancer, those patients go down a different pathway from patients where we say find a KRAS mutation or patients where we do not find a target but have high PD-L1 expression. In lung cancer therapy, it is basically part of everyone's cancer treatment pathway to have this baseline testing done when they have advanced disease.

What are the ongoing research efforts associated with the program?

What I see is that the testing that we do at baseline, we have research coordinators that are linked to clinical trials that Allina is linked with. They are looking to see the results of these tests that we do. They are looking at the patient's diagnosis and stage, and they are trying to match patients based on those genomic test results along with other factors to these clinical trials. We have research coordinators that are looking at everybody closely. They will come forward and approach patients who might be eligible for those clinical trials. It is so important for every patient who has advanced cancer to think about getting on 1 of those trials.

How do you see Allina Health's program evolving in the next few years?

The way I think about genomic testing, and the way I have seen it over time, is that it used to be very expensive and cumbersome, but sort of like our smartphones which have become better and less expensive, most people can afford to have one. I think that is where we are going with genomic testing, too. It is already part of a lot of patients' cancer care, but going forward, eventually everyone's diagnosis is going to have integrated information from genomic testing at the time of diagnosis to determine the classification of the cancer. The pathologist will be thinking that when they issue pathology reports upfront, patients will be matched with the ever-growing choices that are out there for targeted therapy, and they will have a better idea of the biology of their cancer upfront going forward as we start to learn more about cancer biology through genomic testing, and when we are doing it more universally. I think this is only going to keep growing, getting better, and more precise in terms of what we can offer patients for therapy.

What are your predictions for the future of genomic testing in cancer care?

The amount of information we have now is just the tip of the iceberg. Even though it has grown so much since I even started my career from single genes now to being able to do hundreds of genes, most of what we are focused on now is just the sequence of the of the DNA, RNA fusions copy number, a little bit of protein expression, but we are not good at integrating what the profile looks like across a cancer, how those genes interact with one another, how to predict if a patient has a KRAS mutation, an amplification, another gene, how do those predict how a patient will respond to a certain type of treatment? That type of understanding is still in its infancy, and I think what we are going to develop is how those genes and proteins and RNA methylation, how that all interacts, to help us give a more complete picture of cancer to determine what treatment is going to work best. How does that cancer evolve over time with treatment? Where do we pivot and switch to new therapy? I think learning those cut points is going to become really important as we profile cancers through the treatment of a patient.

For the oncologists out there, working with pathologists, radiologists, and pulmonologists to develop a good system for these small specimens [is important] because the information that we are getting out, is based on what is going in. We have to have good quality samples going in to get the good information out. I think the key part of developing a molecular program, whether you have that molecular lab in-house or are sending it out to a commercial lab, is having that good process, preanalytically, to have the best specimens so that you can get the most information out of them. That is usually my message to the oncologists out there. Work with your pathologist, work with the radiologists and pulmonologists, and ask us to make sure you are getting the best specimens.