One of the most critical factors in effecting a viable treatment for debilitating, genetic-related diseases such as cancer is time.
It takes copious amounts of time to sift through the particulars of genome sequencing to identify a driver mutation cell. It takes a substantial amount of time to identify all of the differences between such a cell and a healthy cell. It requires weeks, or possibly months, to match that data with the relevant medical literature to find a medication that can produce a desired result given what is known about a particular person’s cell mutation.
IBM, however, recently unveiled technology that can substantially expedite this process, and possibly offer hope for the future of cancer treatment.
On March 19 the technology company announced a partnership with the New York Genome Center (NYGC) in which it will utilize a prototype of its cognitive computing system, Watson, in a clinical research study for approximately 25 participants with a form of brain cancer known as glioblastoma.
The Watson prototype – which will be delivered to NYGC via the Cloud—was specifically created for genomic research and will be used to expedite personalized treatment to patients. It will apply analytics to Big Data sets of genetic information and medical literature to determine the most effective means of treatment. According to Steve Harvey, Worldwide Leader for Technology and Analytics at IBM:
“DNA and medical literature is the Big Data of biology. One strand of DNA has six billion symbols in it. When you’re trying to compare what a normal cell looks like to what a cancer cell looks like, finding the differences is a complicated process. And even once that’s done, which could take weeks, you’re still left with trying to go through and look at what those differences are. A lot of progress hasn’t been made just because of the sheer volumes of information and the time it takes to process through it.”
Watson, however, can considerably accelerate what was previously a predominantly manual process through its Natural Language Processing (NLP), which puts it at the forefront of the cognitive computing movement. Its NLP capabilities (provided by a complex set of algorithms) effectively make it able to “learn” in ways similar to humans, which helps it to analyze information faster.
In particular, Watson’s NLP capacity will enable it to increasingly understand how the pathways of the body work – which is essential for biological and genomic research. In addition to parsing through the nearly limitless number of combinations of genetic information pertaining to germane cells, it’s celerity can also greatly reduce the time spent going through medical literature regarding treatment options. Harvey commented on the velocity at which Watson can ingest and learn from Big Data.
“It will take the differences between a normal cell and a cancer cell that’s been determined through a sequencing process. Today, once you get those differences, it takes up to three weeks or longer to really go through it and try to identify what the driver mutations are and what drugs may exist to treat those. We can say very strongly in our case that we think we can get that cycle of three weeks or longer literally down to minutes.”
NYGC is a non-profit consortium of some of the most capable scientific, medical, and technological institutions in New York including New York University, Columbia, Cornell, and New York-Presbyterian Hospital, among others. The specific clinical study that Watson will be used for will begin with patients receiving standards of care treatment that are fairly uniform for individuals with glioblastoma. Simultaneously, the respective hospitals that are treating them will send biopsies of their cells (both healthy and otherwise) to NYGC for genome sequencing.
Once that process is accomplished with the aid of Watson and the medical literature is analyzed to indicate that a patient may benefit from a particular treatment such as a specific drug, two things must take place before that treatment is administered. Patients must demonstrate that they are not responding to the typical standards of care; and, their oncologists and a review board must agree on the fact that there is substantial evidence that indicates a patient can benefit from an alternative treatment. If both of those circumstances are present, oncologists will consider implementing treatment based on data provided by Watson.
In addition to developing the Watson prototype for genome sequencing, IBM representatives will continue to work with NYGC on this clinical research study until at least the end of the year. According to Harvey, IBM researchers will be looking for ways to apply the technology across other areas of the NYGC consortium, as well as to other types of cancers. IBM was selected by NYGC as its First Technology Partner, indicating how much the consortium will rely upon the research and technology company for its technological needs.
Perhaps most importantly, Watson’s technology can actually better the way that counteracting cancer is pursued. Enough progress in Watson’s ability to identify a driver mutation cell can actually alter the focus of treatment from where cancer is located to how it originates. Harvey reflected on this possibility:
“For 150 years cancer’s been categorized by where it shows up in the body. But what researchers are finding, and the NIH just completed a study on this a few months ago, is that the exact same mutations are showing up in different parts of the body. That’s leading to the question in the field of genomics that what’s more cost effective to do instead of treating cancer where it’s showing up in the body is, if we understood what was actually causing it, would we treat it differently?”
Harvey’s question is the key difference in the work that Watson is doing at NYGC and that which it is doing at Memorial Sloan-Kettering Cancer Center (MSKCC), where the technology has been used to abet oncologists in utilizing evidence-based treatment for lung cancer. At MSKCC (one of the many entities that is part of the NYGC consortium) researchers are employing Watson’s learning capacity to “teach it the last three decades’ worth of clinical knowledge that’s been developed in how to treat cancer,” Harvey said. The historical approach used at MSKCC is in direct contrast to what Harvey refers to as the “emerging science of genomics” in which Watson is deployed at NYGC.
The Watson ‘ecosystem’ is vast, expands well beyond healthcare research, and encompasses such disparate aspects from consumerization and enterprise uses to cooking. IBM recently allocated a separate business unit, the Watson Group, funds exceeding a billion dollars, approximately $100,000 of which is set aside as venture capital to help derive the creation of forthcoming Watson technology and applications to further its usage and viability in the technological needs of today.
Big Data in the Cloud
The interaction between NYGC and IBM’s Watson is yet another example of the boons that accessing Big Data through the Cloud provides. The Cloud deployment enables researchers at NYGC access to Watson’s full capabilities with limited demands on the former’s infrastructure and existing architecture, with a speed and propensity for analytics that would be difficult (and quite costly) to match onsite.
Ultimately, the partnership between Watson, NYGC, and other medical entities represents a collaborative approach to utilizing technology to eradicate some of humanity’s biggest obstacles. Harvey considered this trend:
“One thing that’s very common when you read about cancer in general in the press is everyone sort of points to the fact that we need more collaborative relationships. We need the people who really understand the biology of the cancer and bring that together with the technology that can get us new ways to approach it. Here you have a highly respected consortium coming together with a highly respected technology company working on a real life project that hopefully will lead to saving lives.”