cancer genomics

It’s the question of the moment Are we living in the age of AI? Or is it still just hype?

When it comes to the latest research in immuno therapy, computational modeling is helping to answer key open questions, such as which patients might respond to which drugs.

"If you were to ask me last year about deep learning, I would probably say, aaah, most of the algorithms that are published are not really answering the important questions yet. But I think this year I am converted. We are starting to use deep learning, and we are starting to see interesting results.”

Shirley Liu is a Professor of Biostatistics at Harvard and the Co-Director of the Center for Functional Cancer Epigenetics at the Dana Farber Cancer Institute. Her lab has very recently put out three algorithms, TRUST, TIMER, and TIDE which represent some very exciting ways that bioinformatics is empowering not only cancer research but treatment decisions.

As a computational biologist, Shirley has found herself highly in demand today as the latest genomic tools such as single cell sequencing generate new amounts of data and as public databases such as TCGA make their rich cohorts available.

In today’s interview, she details these three new algorithms and makes the case that computational modeling has arrived for cancer genomics.

It’s a question we’ve asked on the program before. Are we over relying on the genomics route getting us to biomedical research paradise? Should we be putting more eggs in other baskets?

After combing through lots of clinical trials data, Tony Letai of Dana Farber and the Broad, found that a majority of cancer patients have not benefited from precision medicine. On today’s show he says we need to rethink our approach to cancer research and treatment.

“I think we have a block in our minds in cancer biology about the rules--there are some rules we’re playing by that I don't think we need to play by. I think we can cheat,” he says.

Tony says one of these “unspoken” rules is that we need to use "initial conditions” got from a cancer cell that has been biopsied and killed and broken down for it’s parts, particularly for its “bag of DNA.” Today Tony advocates for an additional approach to genomics, the revival of an older tool, that of screening the live cancer cells against all available drugs. He calls this “functional precision medicine” or a combinational approach, and believes that in 10 years these functional assays will be standard of care in labs everywhere.

Recently we did a show with the CEO of Karius who is bringing sequencing to the world of infectious diseases. What Tony wants to do is bring more of the world of microbiology and cell culture from infectious disease over to the world of cancer treatment.

We have heard of others doing this, as Tony acknowledges. A couple years back we featured Krister Wennerberg from FIMM and last year a private company in Seattle doing something similar. Tony says he has founded a new society to provide network support for this new group called the SFPM or Society for Fuctional Precision Medicine. We finish up at 27 minutes.

A few years ago they were the new kid on the block, and now they are a leader of mainstream genetic testing. Last year their revenue and profit were significantly up.

And yet in today’s interview, CEO Sean George does not sound like someone sitting back on his laurels. In fact, he says his business model has got to change.

We expected Sean to push back against the recent approval of 23andMe’s direct-to-consumer BRCA test and the recent boldness of the DTC space in general, but he surprised us with a welcoming attitude toward his “coopetition.” He says that the DTC companies are providing general education in genetics to the public.

“It’s tailwind for us.”

When asked if they would ever consider going direct to consumer, Sean said that Invitae has already been piloting a broad comprehensive genetic profile for healthy people and has plans to launch it to the general public in the future. All of their tests, he emphasizes however, must be ordered through physicians.

Sean goes on to lay out his vision for the business model of the future.

"The model of single $3,000-$5,000 tests is clearly dead. It does not work. The companies do not scale. The value is there. It’s disappointing to me that it doesn’t work. But people are not paying for it. So our view is you've got to go at it the other way. Let’s unlock it. Let’s make it a utility in healthcare.”

After decades on the loose, it’s cool the cops finally caught him. But is it cool how they caught him?

Nathan Pearson and Laura Hercher are back for April’s headlines. AACR had some more good news about Keytruda, and we take a look at the cancer prediction space.

DNA Day, the first cannabis based therapy to be recommended for FDA approval, the Zuckerberg hearing—there’s lots in here today.

Hanlee Ji is the Senior Associate Director of the Stanford Genome Technology Center as well as an oncologist at Stanford. He’s also a clinical geneticist. In other words, he doesn’t need to take off his glasses and spin around in a phone booth to be able to do about everything.

“I was in fellowship for a long time,” he says in todays interview.

Long reads have been an important theme in the genomics community of late, and Hanlee’s lab recently developed a new method for isolating long fragments of DNA that rivals the long reads of PacBio and Oxford Nanopore. The new method is the first we’ve featured that uses 10X Genomics’ linked reads. The new method also uses CRISPR and CATCH (a new sample prep system from Sage Science), and because it’s done with digital PCR, it offers the nice advantage of only requiring very small sample sizes.

Applications? Hanlee says he’s most excited to use it to identify 're-arrangements’ such as those in congenital disorders or oncogenic drivers.

Hanlee’s lab is also involved in a new clinical trial using precision cancer vaccines that is pulling him headlong into the immuno therapy space.

With a foot deep in the world of genomic technologies and another foot in the clinic, what does Hanlee the oncologist want to tell the technologists? And what does Hanlee the technologist want to tell his colleagues, the cancer docs?

It’s some big questions, and he takes around 27 min to get around them. Enjoy.

Carla Grandori was for thirty years a cancer researcher most recently at the Fred Hutch in Seattle. She had her personal reasons for working on cancer, she tells us in today’s show. Now, she’s the CEO and founder of SEngine, a startup offering something completely new for oncologists around the country.

Carla says she was motivated to step out of the research lab when two patients came to her personally and asked her, “can you study my cancer?”

“When I heard that second request, a light went off. I had thought maybe in 10 years the research would be useful. But then this patient made me think, ‘no, I think we can help now.' When I realized the potential of our technology to help patients immediately, I saw no more boundaries. I said, 'we have to get there.’"

What is this technology? And if it’s so powerful, why hasn’t it been offered before?

SEngine's (think Search Engine) new test, called PARIS, screens individual patient tumor cells onsite against a complete library of over a hundred and fifty cancer drugs. Before now with say, Foundation Medicine, there has been much progress in sequencing tumors and offering lots of omics data virtually. But the PARIS test goes a step further, screening a patient’s actual cells in real time while also combing through accompanying omics data. Until now we didn’t have the technology to deal with the samples with enough precision and in small enough quantities, nor the necessary robotics and software.

The company received CLIA certification back in June and the test is for sale to oncologists now. The next step for SEngine is to build up a major clinical utility validation.

Immuno oncology is now the dominant topic at Mendelspod. From shows with CEOs presenting new panels of predictive biomarker tests, to the firsts at the FDA with the CAR-T approvals as well as the first approval of a drug (Merck’s Keytruda) based on a common biomarker (MSI) rather than on a tumor type, to scientists discussing rare cells of the immune system, we had more podcasts on this topic than any other in 2017.

Rachel Laing and Olivier Lesueur are partners at Bionest, a global life science consulting firm. They work with companies on both the drug and the diagnostic sides of immuno oncology and are at ease switching back and forth between the two in today's discussion.

Complexity is the key word, says Rachel. Immuno oncology has turned the development of biomarkers on its head. Whereas formerly with a simple companion diagnostic, one starts with a very specific biomarker develops from there.

“With immuno oncology, it’s very different from that,” she says. "You’re not just dealing with a tumor cell and the signaling that goes on there, you’re dealing with the immune system. And there’s a lot of interplay between the tumor, the microenvironment, and the immune system. It’s unrealistic to think you could get away with just measuring one biomarker.”

What are the challenges for diagnostics companies in commercializing various biomarkers into ongoing testing for the same patient, or what’s becoming known as “real time oncology?”

Olivier addresses the great reimbursement irony in the field of drugs and diagnostics. Without the biomarker tests, the drugs may not work for the right patients. Yet the diagnostics companies receive pennies while the drug makers are getting away with— well, we all know that issue. Olivier says one drug maker recently preempted this problem by offering to just pay for the diagnostic up front in geographies where the diagnostic reimbursement was being questioned, which is good for everyone, including the patients, and shows some vision. This problem of undervaluing diagnostics has bedeviled our industry for a long time and becomes especially crucial with immuno therapies that rely on patient stratification and ongoing profiling.

In our final question about what we’ll see in the year ahead, it’s cautioned that though patients are seeing more and new therapy options, the complexity of treatment is becoming ever more demanding on some people who are already pretty sick.

The past few months have seen the first approvals at the FDA for CAR-T cancer therapies. But trials have been going on for years. And not just by big pharma.

Today’s guest, Dr. Michael Jensen, is a researcher at Seattle Children’s Hospital and started his first CAR-T cell trial back in 2002.

“It’s been a long road to get to the point where we understand enough about the science and physiology of this technology to make it work int he clinic," he tells us in today's interview.

Seattle Children’s has set out a tremendously bold vision: think hybrid academic center and biotech company. They are currently building a $350 million, half a million square foot research tower along with a 30,000 square foot GMP manufacturing facility--all under the non-profit umbrella. This is an upgrade on the current GMP facility and will be one of the biggest such facilities of its kind in academia.

Michael says that the center is out ahead of big pharma with CAR-T therapies because they are not going for big drug approvals, but rather focused on early proof of concept work.

“Our role is to innovate in early phase studies with the hope that positive data will result in a Novartis or Gilead or Juno or Bluebird coming in and taking it across the finish line.”

What the non-profit center is able to do that big pharma never will be able to, therefore, is to go after all cancers and pediatric diseases, no matter how rare.

We begin the interview with a rundown on the ongoing CAR-T trials at Seattle Children's and their plans to translate success in blood cancers to solid tumor cancers.

We’ve heard a lot this year about the search for new structural variants and the hope that scientists will find new causal linkages for diseases such as cancer. But will the genome still yield dramatic genetic signatures such as KRAS, BRAF and EGFR that have been so helpful in cancer treatment?

Today’s guest says, yes, and he’s on the trail.

Jim Broach is the Director of Penn State’s Center for Personalized Medicine. He and his team have come up with the highest resolution genomic data to date on certain cancer cell lines using sequencing and mapping tools. In some cell lines his research has revealed 150-200 more structural variants than had previously been discovered.

“There are a whole set of structural variants which haven’t been taken into consideration to date,” he says in today’s interview. "For the next couple of years, this is the dark matter of the cancer genome. We’ve got to sort out which of these structural variants are going to be relevant in understanding how best to treat the patients. Once we generate that information, I think these structural variants will be just as relevant as the point mutations or as large scale translocations."

Jim mentions paired end reads and PacBio’s new long read technology, but the main tool he talks about is Bionano’s optical mapping technology. Previously the field used karyotyping to look for variants of this size, but he says Bionano has got their technology to the quality and price point where it will now replace the older technology.

How will Jim’s research impact treatment in the clinic? He is doing de novo sequences of cancer cell lines. Does he envision the need for de novo sequencing of a patient’s cells as part of a commercial assay?

When we first talked with Tim Triche of LA Children's Hospital, we found out he was a bit of an outlier among cancer researchers. He was an advocate for poking around in the non-coding RNA.

Today we welcome Tim back to the show to talk about a new gene panel that he has designed specifically for childhood cancers. It’s a first of its kind and was modeled quite closely on the gene panel for the NCI’s MATCH trial. The new panel has both a DNA and an RNA component, and the RNA side is by far the biggest.

"There are 1,400 different amplicons on this panel looking for RNA fusions. Thermo Fisher tells me it’s the most ambitious RNA panel that they’ve ever undertaken," Tim says in today's interview.

"When 100 cancer patients walk in your office, then 100 cancer patients walk in your office," says Tim, quoting a common line in the field that points to the uniqueness of every cancer.

Yet even though every cancer is different, certain biological commonalities combined with better sequencing tools is enabling the design of new gene panels to guide in diagnosis and treatment. More and more a cancer is looked at based on the drug that might treat it rather than the organ in which it grows. The new panel can guide this treatment.

Some of the most important targets on the panel are RNA fusion transcripts. What are they, and why are they so important for helping kids?

Childhood cancers come from inherited mutations, whereas most adult cancers have to do with the skin or the linings of the organs due to mutations caused by environmental impacts. Fusion transcripts are very common in the youth cancers and have been a big part of routine diagnostics.

If a mutation is there early in life, is it likely to turn into cancer sooner rather than later? Yes, says Tim.

“If you look at the incidence of childhood tumors, there’s a big bump in the first months or year or two of life, and then they disappear thereafter."

Additional benefits from these new next gen sequencing panels are that they can work with very small “real world” samples of tumor tissue, and they can also be used as discovery tools. Tim says the panel, called OncoKids, is ready to go for frontline therapy, and is hoping to get the word out to oncologists everywhere.