NYU Langone Health and Dana-Farber Cancer Institute have teamed up to develop a homegrown digital oncology tool, designed to help clinicians access and evaluate the latest research, guidelines and therapies for cancer patients.

The organizations piloted the tool, called Solavia Decision Suite, at NYU Langone in June.

With it, oncologists are able to align patient data with a body of relevant clinical research inside the health system’s electronic health record and make treatment decisions accordingly. 

“For oncologists to decide which chemotherapy a patient receives is a complex matter because oncology research changes quite a bit,” said Nader Mherabi, executive vice president, vice dean, and chief digital and information officer at NYU Langone.

The two nonprofits aren’t keeping the technology to themselves. Solavia is now commercially available for other health systems to use — a rare move for hospitals, which usually opt to keep digital tools in-house or buy them from software vendors.

NYU Langone and Dana-Farber are far from the first institutions to develop in-house tools to address needs for specialty care. In March, Geisinger Health System built its own genomic medicine platform and clinical decision support tools centered around precision medicine. 

Of course, not every organization has the financial and operational resources to build a tool from scratch. NYU Langone maintains a technology ventures arm that supports the commercialization of certain tech products.

Mherabi sat down with Healthcare Dive to discuss Solavia, the decision to build rather than buy and what health system leaders should consider when developing their own technology at a time when enterprise software costs are high.

Editor’s note: This interview has been edited for clarity and brevity.

HEALTHCARE DIVE: How did the partnership between NYU Langone and Dana-Farber come to be? What did each institution bring to the table that made this collaboration essential?

NADER MHERABI: We both were using a product beforehand for oncology pathways from Phillips. We discovered that the content in that tool was actually curated by Dana-Farber. So when Philips decided to sunset their oncology business, we were left with the choice of finding a new vendor or partnering with Dana-Farber to do something more robust, meaningful and much more collaborative. 

Dana-Farber came with a lot of oncology informatics and content, and then they found that NYU Langone is advanced in terms of product development and product lifecycle. So that synergy was there. We knew what the pitfalls and pain points of the other product were, and we had good insights into what we could build that would really please our oncologists and be great for patient care.

What was the appeal of developing a tool from scratch rather than adopting an existing one? Why shift from users to developers?

We realized initially that the beauty of creating a product of this variety is that it can be highly integrated into the clinician workflow. Because physicians' time is so limited, information and workflows have to be in front of them. We wanted to make sure it was tightly integrated into our electronic health record, which happened to be Epic for both NYU Langone and Dana-Farber. It doesn't necessarily have to be Epic — this can be integrated with other products such as Cerner or other EHR products, or standalone.

Another draw was the ease of use because oncology pathways can be highly complex. There are multinested branches because you want to find the best treatment for the patient. You want it to be intuitive, extremely fast and responsive.

The other thing is that the product is scalable. It’s cloud-based, and it was important to make sure that we have the latest modern piece of software that's highly functional and extendable.

You made the choice to launch Solavia internally at NYU Langone before commercializing it. What was the thinking behind that decision, and what were you trying to learn during that internal pilot phase?

One, we needed to replace the software. We wanted to do that for our own clinicians and patients. Second, we have an incredible cancer center, and so does Dana-Farber. So between all the people we work with, we could gather a lot of input to make sure the product makes sense and meets clinicians' needs. We had an open door between cancer center oncologists at Dana-Farber and NYU. We did a lot of discovery with them when we built the product. Of course, we wanted to launch it for ourselves, and then prove to other health systems that it truly works, and they can come and actually see a working product in our institution.

What does your governance structure look like, and how does it ensure both institutions maintain alignment as the platform scales?

We have a governance board and official agreements between the two parties and we have our legal, compliance, ventures teams — they all engage. We have a good governance structure to make sure the products not only support us and Dana-Farber but also other colleagues who are going to join and use the product. It's truly designed to be multitenant, where the data of every institution is separately handled. We've gone through a lot of security reviews. It’s the same thing we do when we buy a product. We have to go through all of that ourselves to make sure that it's truly a secure product and it's highly segregated within each institution.

Do you think we will see more healthcare organizations choosing to build platforms rather than adopting general-purpose tools, particularly in specialty care? What do health system leaders need to consider before making that choice?

Institutions often buy a big stack of software where they use maybe 30% of the software but they’re still paying for 100%. So that's a concern. Also, with the advances of amazing technology, cloud-based capabilities and AI, the timeline for product development and building something that specifically solves a problem has matured.

I don't think every other health system is going to build all their own software. But there is some niche software that I think institutions sometimes have to weigh the options if they have that capability, which not everyone does. But if you do, it becomes a decision point. The fact is software development is becoming a lot easier with AI and advanced tools in the cloud, and SaaS products tend to offer a lot more than you really want or need, and they are expensive. But every institution has to weigh the options. It’s not one-size-fits-all