NEW YORK – A new startup called Stitch Bio is combining whole-genome sequencing, CRISPR technology, liquid biopsy, and gene therapy to target cancer cells with gene fusions in order to make them druggable. The company believes its approach is both applicable to a wide variety of cancers and highly precise and individualized.
Stitch Bio is the newest venture of Anthony Shuber, former chief technology officer of Exact Sciences and current president and chief operating officer of Genetics Research, which he cofounded in 2018 to commercialize a CRISPR-based sample prep method called "negative enrichment technology."
Now, Shuber is using his CRISPR experience and his knowledge of cancer testing from helping to develop Exact's hugely successful stool-based colorectal cancer diagnostic test Cologuard, to launch a company that he hopes will turn cancer from a deadly disease to a chronic one.
"Right now, there are a lot of companies out there developing liquid biopsy tests, and they're able to tell patients they're recurring before [their cancer is] picked up by MRI or CT scan, but at this point in time, there's really nothing one can do for these patients other than just simply try again with the drugs that currently exist," Shuber said. "What makes Stitch Bio unique is that we have a foundation of proprietary ideas not only for detecting recurrence and resistance of cancer patients who are undergoing primary treatment, but if a patient does recur or becomes resistant to a primary treatment, we can actually do something about it."
The idea, he said, is to specifically look for gene fusions. The company would collect a sample of a patient's primary tumor and analyze it with whole-genome or whole-exome sequencing. Once gene fusions are determined, CRISPR guide RNAs (gRNA) would be designed to target cells with those fusions and manipulate them in some way to make them druggable with currently available therapies. Multiple gRNAs could be applied at once to attack the tumor cells in multiple ways to cut off the cancer cells' resistance to therapy, and liquid biopsy would be used to monitor patients' recurrence or resistance rates.
"Today, everybody's using CRISPR on an in vivo basis to address single-gene disorders and repair genes in diseases that are deficient in certain proteins, or delete genes where their gene product is directly responsible for disease," Shuber said. "In our case, we're repurposing CRISPR to be able to manipulate the fusions that are very specific to cancer, that are not present in normal cells, and be able to make those fusions druggable. So, we are going to find the diagnostic biomarkers that are specific to a person's lesion, recurrence, or resistance and be able to target that to enable already existing drugs to treat that cancer."
The technique of using CRISPR to create druggable targets is somewhat similar to a method being explored at ChristianaCare's Gene Editing Institute, where institute Director Eric Kmiec and his colleagues have been working to combine CRISPR genome editing and chemotherapy in non-small cell lung cancer patients in order to improve the efficacy of certain chemotherapy agents.
However, the difference is that Kmiec and his team are using CRISPR to knock down the NRF2 gene to render patients more sensitive to chemotherapeutic agents such as cisplatin and carboplatin, whereas Stitch Bio's method is focused on the manipulation of gene fusions.
There are several possibilities for how to achieve that manipulation. One example stems from the work of University of Pittsburgh professor Jianhua Luo, who sits on Stitch Bio's scientific advisory board. In 2017, Luo was senior author on a Nature Biotechnology study in which he and his colleagues demonstrated that CRISPR-Cas9 could be used to introduce the gene encoding the prodrug-converting enzyme herpes simplex virus type 1 thymidine kinase into the genomes of cancer cells carrying unique sequences resulting from genome rearrangements. Specifically, they targeted the breakpoints of TMEM135-CCDC67 and MAN2A1-FER fusions in human prostate cancer or hepatocellular carcinoma cells with a Cas9 construct in a way that made the cancer cells vulnerable to treatment with the anti- cytomegalovirus medication ganciclovir.
"We're not editing those sequences to either repair them or delete them," Shuber explained. "We're using CRISPR to edit the unique two sequences surrounding the fusion to be able to insert whatever we want to. In [Luo's] case, he inserted thymidine kinase. We could insert a short peptide that gets expressed on the cell surface of the tumor and subsequently could be used as a synthetic neoantigen to be targeted by immunotherapy. Those are just two examples."
He noted that the company has filed for patent protection on several unique treatment approaches.
John Quackenbush, professor of computational biology and bioinformatics and chair of the biostatistics department at the Harvard T.H. Chan School of Public Health, believes that Stitch Bio's method adds a new dimension to targeted therapy.
The idea behind targeted therapy is to find something that was distinct and unique to the tumor cell that wasn't in normal cells, he said. But tumor cells also produce a plethora of genomic and genetic changes, some of which are fusions events, that can also be targeted. In some cancers, like leukemia, the fusions are the drivers that trigger the growth of the disease. In others, the fusions have a modifying effect that comes into play later on in the tumor's development. But in most cases, there are fusion events that are unique to individual cancers, and that create a target for treatment while sparing normal cells.
"So now you're going to have a big target you can hit with CRISPR and you can insert all sorts of things," Quackenbush said. "You can imagine inserting antigens which are really specific, which don't even have to be human antigens, that can be targeted by antibodies that will carry chemotherapy directly to those cells. Or you can imagine incorporating a suicide gene that will drive the cell to death and extinction. There are a whole host of different approaches you can imagine if you can specifically plug a synthetic construct into tumor cells, and only tumor cells, to allow you to target those tumors and destroy them."
Importantly, Shuber said, the method is both applicable to just about any type of cancer that develops gene fusions, and also highly precise in its targeting of a specific fusion signature. The process would be the same in each patient, he said, with the only difference between cancers or stages being the guide RNAs associated with the Cas9 approach.
"It's precision in the sense that if you were a patient, I would need to know what fusion events your tumors carries, and I'd want to know as many of those as possible," Quackenbush further added. "This could potentially target multiple fusion events, but I could do it in any tissue as long as a fusion event exists in that tissue."
Further, the emphasis on gene fusions could position Stitch Bio's approach as a precision oncology treatment alternative for pediatric cancers, which are largely driven by such fusions.
It's currently still early days for the company, which is currently raising capital in order to develop its technology and design appropriate research studies. It aims to conduct its preclinical studies over the next couple of years, Shuber said, and is in discussions with a contract research organization to conduct experimentation on different cancers and different fusions.
Pediatric cancer will be one area of interest, he noted. And as Stitch Bio goes further into recurrence and resistance, lung cancer, liver cancer, and ovarian cancer will likely feature very high on its list of priorities, as well.
Quackenbush concurred, noting that pediatric cancers and leukemias are a good starting point.
"If you wanted to demonstrate that a therapy is efficacious, you want a place where the targets are pretty clean, and you can actually show that you can improve over existing therapies," he said. "A lot of leukemias are driven by simple gene fusions. In multiple myeloma, there are fusions between chromosome 4 and chromosome 14 that are important in a subset of the disease. So, I think you can find a few places where there are really clear fusion events that are distinct and common. Those are great targets, because in many cases they're driving the disease, and we also know that we can find enough patients to actually demonstrate that this application is efficacious."
He also added that Stitch Bio could have a big impact by concentrating at first on demonstrating the efficacy of its technology in cancers with high fatality rates and few effective treatment options, like pancreatic cancer and ovarian cancer.
Notably, the company is also planning to build a database of cancer-related gene fusions, so it can learn more about the biology of these rearrangements and how they affect tumor development. From a business perspective, Shuber used Foundation Medicine as an example of what he wants to achieve with Stitch Bio.
"Where Foundation is sequencing primary tumors to find drivers that inform therapeutic players, we would sequence primary tumors and develop a database of fusions, while at the same time having identified patient-specific fusion signatures for any future recurrence," he said. "And then, if someone had a recurrence, we would inform the physician of a fusion signature and let them know that we have that data."
Foundation Medicine's FoundationCore database contains more than 400,000 patient profiles, and typically adds more than 2,000 samples per week. The pan-tumor knowledgebase is fed from a variety of the company's cancer assays.
According to Quackenbush, the Stitch Bio database will be an invaluable source for knowledge on the basic biology of gene fusions in cancer. "The surveys of fusion events in tumors is not as extensive as [those of] single-base mutations," he said. "I really like this distinct approach to looking at fusions as a target for therapy, even if they aren't functional."
He also noted that Stitch Bio's ambitions to become a precision medicine company may rely heavily on its success at collecting data. "Any precision medicine company today … has to be, at least in part, if not at its core, a data company," Quackenbush said, "Data is what's going to drive approval of these drugs. It's also going to drive the development of the next generation [of technology]."