By Allison Proffitt
September 9, 2015 | It’s one of her most consistent talking points. Mary-Claire King, the 2014 Lasker Award-winning scientist who discovered the BRCA1 gene implicated in breast cancer, has issued several public challenges to make BRCA1 and BRCA2 testing available to all 30-year-old women.
Last October, King made her case at the HudsonAlpha Institute for Biotechnology in Huntsville, Alabama, where she spoke at the annual breast and ovarian cancer awareness event and fundraiser.
HudsonAlpha has accepted the challenge.
Starting on October 30, 2015, every 30-year-old woman in Huntsville and Madison County, Alabama, will be eligible for a free cancer screening panel thanks to the Information is Power initiative of the HudsonAlpha Institute. Kailos Genetics, a six-year-old diagnostics company that spun out of HudsonAlpha and is still headquartered in the same building, is partnering with HudsonAlpha on the initiative and will conduct the testing using its 23-gene Women’s Health Panel.
Women who are 30 years old at any point between October 29, 2015 and October 28, 2016 and live in Madison County, Alabama, are eligible to participate. Individuals older than 17—male or female—and living in Madison, Morgan, Limestone, Marshall, or Jackson County, Alabama, qualify for a $99 discounted rate on the Women’s Health panel.
The initiative is a bold step forward for population-wide genetic testing. While King and others have called for such testing, some researchers and ethicists have warned that we are not yet ready to take that step.
Kimberly Strong, director of the Ethics and Genomics Program at HudsonAlpha and the faculty member leading the initiative, has heard the objections before. “I acknowledge that an initiative like this is not universally accepted as something that should be done at this moment in time,” she said. Currently, genetic testing for genes linked to breast and ovarian cancer is offered to individuals who have already been diagnosed with cancer, or are from a family that is considered high risk. “But these types of testing scenarios do miss out on identifying some people who are unknowingly at increased risk,” Strong said. “Our initiative aims to conduct larger population-level testing to raise awareness of genetic and traditional screening methods and identify additional individuals at increased cancer risk who may not have had access to testing through other avenues.”
Identifying patients who are at increased risk will allow their healthcare providers to create a new screening plan, Strong said. “Genetic testing never takes the place of traditional screening, of mammograms and colonoscopies,” she emphasized. “What it does do is give another piece of the puzzle as to what might be going on, what someone’s risk might be.”
But risk is a tricky thing to understand.
Matt Yurgelun, an oncologist at Dana-Farber Cancer Institute and Harvard Medical School, recently published an article in the Journal of Clinical Oncology (doi: 10.1200/JCO.2015.60.8596) warning of some of the risks associated with population-wide BRCA1/2 screening.
Yurgelun is not familiar with HudsonAlpha’s plan, and so declined to speak about the specifics, but he called population-wide screening projects “a very laudable idea.”
“It’s been a great talking point to get discussion about genetic medicine into the mainstream public discourse, finding ways to improve access to genetic testing, and improve the public awareness of genetic testing,” Yurgelun said. “At the same time, there are aspects of genetic testing that need to be kept in mind when you go from using it in a high-risk setting, which is how it’s classically used now, to using it as a screening test for the general population.
Yurgelun and his coauthors highlighted variants of unknown significance, genetic counseling, and discrimination as hazards on the way to population-wide screening.
“Population-based BRCA1/2 screening would be intrinsically different from high-risk genetic testing in ways that will likely accentuate the limitations, particularly for minority women, while also reducing potential benefits,” the authors wrote. “Potential effects of false-negative results, including the possibility of false reassurance that might lead some women to forego appropriate care, are of concern. Although professional genetic counseling can mitigate these risks, the volume generated by population-wide BRCA1/2 screening would mandate that traditional informed consent and counseling models be completely overhauled.”
Counseling at Scale
HudsonAlpha is taking the need for genetic counseling seriously, while exploring ways to do it at scale. Participants in the HudsonAlpha initiative will begin the process by visiting the Information is Power website and ordering a test kit (once the program launches on October 30). In order to proceed, people will watch a video outlining how genetic testing works and what it can—and can’t—tell you.
“We are trying to simulate some of the traditional things that occur, so that at the very least the topics that are discussed in a traditional genetic counseling process would be included in that pre-test video,” Strong said. “It will give a description of what a test can tell someone, and probably more importantly, what it can’t tell them. [A test is] not going to diagnose cancer, for instance. Even a positive result for one of these gene changes doesn’t equal cancer, and a negative result doesn’t mean one would never get cancer.”
The video will explore what one’s test results might mean for biological family members and touch on GINA—the Genetic Information Nondiscrimination Act—highlighting that GINA does not provide protection for life, disability, or long term care insurance. Each section of the video will be short—just a handful of minutes per topic, Strong said. She’s not sure how long the entire video will be when it’s completed.
Importantly, the Information is Power initiative is not a clinical or research study. There has been no IRB approval, and Strong and HudsonAlpha are not collecting data on the participants or the process. While everyone who receives the test must check a box stating that they have watched the video to proceed, this isn’t clinical informed consent. Participants aren’t tested on their understanding, and while they can opt in to an email list for a possible future study, no such study is currently planned.
After watching the video, participants move forward with testing. And at this point, the initiative shifts into Kailos’ domain.
A Dissatisfying Model
Kailos Genetics’ Women’s Health Panel screens for 23 genes associated with breast, ovarian, uterine, colon, and pancreatic cancers, including BRCA1/2, PALB2, TP53, and more. For Kailos, the initiative launch will come just months after the company started offering its genetic tests for sale online to patients.
 |
| Troy Moore |
The consumer-facing product marks the next step in an ongoing business model evolution, Troy Moore, Kailos’ Chief Scientific Officer, told me during a visit to Kailos headquarters. The company’s foundational technology is the product of a HudsonAlpha Ph.D. student. It was licensed into the company and offered first to research and academic labs, before the company morphed into a more traditional diagnostics company business. But after just a few years, Moore said, Kailos was “somewhat dissatisfied with that model.”
Late in 2014, the company decided to no longer offer testing for insurance reimbursement and instead focus on a consumer-directed approach. “We would remain a CLIA-certified laboratory, work on our CAP certification, our New York certification, all that good stuff—still bring that level of testing—but allow consumers to order the tests from us directly,” Moore explained.
Kailos currently offers a menu of pharmacogenomics tests exploring the genetics related to patient response to antidepressants, tamoxifen, pain management medications, Plavix, oral contraceptives, and stomach acid reducers. The tests are regularly $149 to $299 each, but are currently listed at $99 each.
More tests are being announced regularly, said Randy Bachmeyer, Kailos CTO. A cardiology test is coming soon, and several risk-based panels will be introduced, including a wider release for the women’s health panel as well as men’s health and eye health panels.
Kailos isn’t the only company anxious to cut out the insurance middleman. Invitae, Color Genomics, and Pathway Genomics all announced products this year that are sold through variations on the same approach. In Kailos’ case, when a customer orders a test, she types in the name of her physician who should receive the test results. Kailos faxes that doctor an explanation of the test and the requisition.
“Several of them come back to us and say, ‘I’m not the appropriate physician for this. They are asking about Plavix testing and I’m their GP. It really should be their cardiologist,’” Moore explained. “In which case, we’ll reach out to the cardiologist or go back to the patient and explain the situation. If they don’t have a physician available to them, or one that will order the test for them, then our medical director will look it over and sign off on the ordering of the test.”
The process seems a bit cumbersome compared, for example, to Color’s model of instantly connecting customers to an ordering physician through their online store. Moore acknowledges that the approach might slow them down a bit, but he’s unconcerned.
“That’s okay. Part of the goal here is physician education, for them to know this type of testing is available and it’s available without the normal hassles of reimbursement.”
 |
| Randy Bachmeyer |
Kailos views the physician’s role as very important, explained Bachmeyer. “It’s [the patient’s] physician that either has put them on drugs, or will put them on drugs, so if we can’t engage that patient or that customer’s physician, we’ve left it all up to him to talk to his doctor. We want him to be able to talk to his doctor—we want him to have a report to look at as well—but it’s very important that he works with his primary physician.”
“If their physician won’t take [the reports] seriously, I hope they look for another physician,” Moore added. “They should take it seriously.”
The same requisition process will be in place for the Information is Power initiative, and both Kailos and HudsonAlpha are working on educational outreach to doctors locally. “We’re doing active outreach to physicians in the area,” Strong told me. “We’re having face-to-face meetings. Lots of emails are going out.”
“OBGYNs understand BRCA1 and BRCA2 testing and its relationship to breast cancer and ovarian cancer, but they’re not all comfortable with a list of 22 other genes and how that will impact them,” Moore says. “Until it’s put out there, and they have to deal with it, [adoption] is going to be really, really slow, which means people aren’t necessarily going to get the best healthcare they could get.”
The Information is Power initiative is meant to force their hand.
Sample Progress
While doctors are signing requisitions, the patient’s sample moves through the normal Kailos pipeline.
All Kailos customers receive cheek swab kits in the mail and return them to Kailos’ laboratory in the HudsonAlpha building. Samples undergo targeted sequencing, a hybridization method with limited PCR. Panels are designed on Kailos Blue, the company’s design-to-reports platform. The design process was honed during the company’s years serving research customers, explained Bachmeyer. “We’ve probably done over 100 different panels for different research customers.”
To precisely target genes, Kailos Blue suggests where to put down primers from a reservoir of about 4,000 options, Bachmeyer said. “When we produce the panel for the very first design, we get probably about a 90%-95% success rate. Of course we are doing clinical testing, and that’s not good enough. That’s where you can really go in and start tweaking the individual design until you’ve got the 100% you need to do that particular test.”
Sequencing happens in-house on one of five Illumina MiSeqs or one NextSeq instrument. Kailos Blue includes its own LIMS system to move samples through the process. Analytics run on the Amazon Cloud using a core pipeline of BWA and GATK, modified for targeted sequencing, Bachmeyer said. After 25-30 quality checks on each sample, results are stored in Amazon S3.
At this point, if the requisition paperwork is not in order, the test and its results are discarded (and paying customers get a refund). If the requisition paperwork is in, reports are generated. Pharmacogenomic panels are reported based on PharmGKB guidelines; oncology reports rely on ClinVar. Test results are returned in both a patient-facing report and a physician-facing report.
For Kailos’ pharmacogenomics tests, the practical results might be specific dosing recommendations. The patient might be told that their test results warrant discussing a decreased medication dose with their doctor; the physician report might recommend cutting the dose by a more-specific 50%. Both reports were developed specifically for their intended readers with an emphasis on usability and clarity, Moore explains.
For the HudsonAlpha Information is Power initiative, any panels positive for mutations will be returned to a genetic counselor and the ordering physician several days before reports are sent to the patient. The genetic counselor will touch base with the physician first, then reach out to the patient.
“We want to give the physician the opportunity to be the one to contact the patient,” Strong explained. “But regardless of whether the physician has contacted the patient or not, the genetic counselor will be contacting the patient.”
The Canadian Project
HudsonAlpha calls the project groundbreaking, but as I sat in the Kailos conference room, the whole thing sounded vaguely familiar.
Mohammad Akbari, an assistant professor in the Dalla Lana School of Public Health at the University of Toronto, presented plans for a similar project at The Clinical Genome Conference* in June of last year. I reached out to see how that effort was progressing.
Akbari, too, advocates for universal screening.
“Right now, we’re seeing people waiting until they are diagnosed with breast or ovarian cancer… and completing their [first] course of treatment—a year or two after their diagnosis!—[before they have genetic testing],” Akbari said. “The other problem is that when offering genetic testing we use so many different criteria: their family history, having a certain number of breast or ovarian cancer cases in their family, age of diagnosis. It’s believed with all those different criteria we are still missing a significant proportion of the carriers.”
A study from Akbari’s group supports the theory (doi: 10.1038/bjc.2013.309). Researchers at the University of Toronto screened for BRCA mutations in about 6,000 Jewish women in Ontario from 2008 to 2012 who did not otherwise qualify for BRCA screening. “Of those 6,000, we found more than 1% of them—93—had a mutation,” he said.
Those findings prompted a larger vision: the Universal Population-Based BRCAness Screening project. Akbari and his colleagues wanted to target women 30 to 60 years old in Ontario, Canada—about three million women. They planned to offer BRCA1, BRCA2, and PALB2 testing and had partnered with WaferGen for target enrichment.
That was the plan Akbari presented at TCGC a year ago. “Because of so many different reasons, we were not successful in implementing it,” Akbari told me last week when I checked in for an update. “But we are still working on it.”
The major challenges are funding and regulations around direct-to-consumer genetic testing in Canada. Akbari’s hope now is to collaborate with one of the large hospitals in Toronto that is already offering genetic testing and launch universal testing in the first or second quarter of 2016.
Then he’ll be left with only the funding issue.
“We have been working in the last couple of years to find a source of funding to cover such genetic testing,” he said. “Unfortunately, so far we have not secured any funding for it.”
Akbari hopes to offer the testing for about $150-$230, and to educate and advocate for women to buy the test themselves. That, too, will be a challenge. “Especially here in Canada, people are so used to not paying for their health services, it’s really hard to convince people to pay for such a thing,” he said.
In the meantime, Akbari has been modeling cancer in Ontario. “We showed in our modeling of Ontario's population that if we can do population-based genetic testing for BRCA1 and BRCA2, we can decrease the incidence of breast and ovarian cancer significantly in the population.” (A paper with more details has been accepted for publication.)
With enough data points, Akbari hopes to prove that genetic screening should be universal, and he hopes eventually such tests will be covered by the Canadian healthcare system. “Our ethical responsibility as a group of academic scientists is to show the benefit of such population-based genetic testing,” he said.
The Numbers Game
HudsonAlpha’s initiative might overwhelm its organizers as well, but the smaller, better funded project has fewer obstacles to success.
HudsonAlpha has calculated that there are 1,500 women who would qualify for free screening in the Huntsville area covered by Information is Power. In the surrounding counties, there might be about 600,000 individuals (estimated from 2013 numbers) eligible for the discounted rate of $99 per test. That’s a lot of tests, but far fewer than Akbari’s three million.
Kailos is donating the testing for 30-year-old women and offering the discounted price to the rest of the eligible participants. The lab can currently run about 2,000 samples a month with four technicians, a testament to the ease of the protocol, said Bachmeyer.
There are also various overhead costs: the website, the educational materials, and the genetic counselors’ time. HudsonAlpha is assuming those costs, explained Maureen Mack, Senior Director for Marketing and Communications at HudsonAlpha.
There are several genetic counselors committed to the project thus far—three from HudsonAlpha and two more that can be contracted in if needed, said Moore. Currently, genetic counseling will be offered to everyone whose tests reveal mutations. Strong declined to predict how many people that would be, but did say, “We are expecting numbers to be small.” For non-Ashkenazi populations, studies have estimated that 1 in 400 women carries a BRCA mutation. Troy Moore expects the rates for the other genes in the Kailos panel to be even lower.
The project will kick off at the Tie the Ribbon fundraiser on October 29. Everyone who attends the luncheon will receive a kit, though not everyone will be eligible for free screening. Tickets to Tie the Ribbon are $60 each. Once the initiative has launched, HudsonAlpha will be doing community outreach and education. “Evaluation of the initiative and community consultation is going to occur over the coming year,” said Strong.
Mack didn’t share a budget for the project, but said HudsonAlpha has been fundraising since last October, and the cities of Huntsville and Madison, and Madison County have also committed funds.
We are committed to every woman who is eligible, Mack said. “Would we be surprised if 1,500 women took advantage of it? Maybe. But we will keep our commitment.”
* The Clinical Genome Conference—TCGC—is an annual event held each June in San Francisco and run by Cambridge Healthtech Institute, the parent company for Diagnostics World News.