Dr Dubinsky reviews her presentation at Advances in Inflammatory Bowel Diseases 2021 on the progress toward and future promise of personalized care for patients with IBD.

Marla Dubinsky, MD, is codirector of the Feinstein IBD Center at the Icahn School of Medicine at Mt Sinai in New York, New York.

TRANSCRIPT:

Dr. Marla Dubinsky:  Hello. I'm Dr. Marla Dubinsky from the Icahn School of Medicine. I'm delighted to share with you some high-level ideas and points that came out of the lecture that I just gave here at the Advances in IBD meeting in Orlando. My topic today was personalized medicine, yesterday, today, and tomorrow.

Big topic to cover in a short period of time, but I want to run you through some key ideas that we shared today. It's important to recognize that we hold oncology as our North Star of, can we get to a point where we can personalize our therapies to the patient? There's a couple of key things that I shared with the audience today.

The first thing is to think about, do we have the tools to select the right patient to be getting the therapies that we deem most effective? Is it possible that some patients will have a different prognosis, and take a different trajectory or a different disease course where we can use different therapies based on prognosis?

Right now, we take all IBD patients, assume they have the same natural history and trajectory, and then apply therapies based on drugs we've used for a long time, or drugs that we think are the most effective. What we need to start doing is deconstructing that and saying, "All right, is the future a more biologic approach?"

Am I going to be able to use prognostic markers, decide, "Is this someone who can take a little bit more time, which we used to call the step-up approach, and I can use that for now?" or is this a patient that we're going to go right to our most effective therapies, which used to be called the top-down approach?

To be truthful, when do we ever go in with our most effective therapy and then say, "You know what? I think you're ready to de-escalate." I'm not so sure we're there yet, and we need more data to help us decide whether there is such a thing as top-down.

We think about the right patient, which is really about predicting prognosis. It has been both clinical factors that have been put out there and published. It's been serologies, which is what I've been most interested in along with my other colleagues when I was at Cedars-Sinai. It was the idea of, "Are there serologic immune markers that are present that predict prognosis?" We've shown for a while now that, yes, some of the antimicrobial-based antibodies are present in people who complicate in advance of that complication. We've shown that in a number of cohorts.

The most recent 1 or 2 that are relevant is the RISK cohort, which is our pediatric newly-diagnosed Crohn's cohort, where we show that patients who present with certain serologies at diagnosis predicted prospectively increased rates, particularly of internal penetrating disease.

We've also shown in a cohort called the PREDICTS cohort that was led by Jean-Frederic Colombel and other colleagues, where they were able to see these serologies be present out to 6 to 8 to 10 years before diagnosis.

Those patients who had these biomarkers or antimicrobial antibodies already had a higher likelihood of presenting with a new case of diagnosis of IBD, and being complicated at diagnosis. Serology has been established as more prognostic, less diagnostic markers.

Genetic factors, interestingly, have been suggested that maybe by having these more genome-wise polygenic risk scores, that will give us more of a delineation of whether genetics may predict prognosis -- I'm not sure -- or maybe genetics is purely based on risk of disease development, and that needs a little bit more finessing.

I think that, at the end of the day, one of the things that I shared with the audience is that there's no one biomarker that's going to be the magic bullet to say, "You have this biomarker. You're going to get this prognosis."

At the end of the day, we're going to be developing these risk models -- or risk tools, perhaps -- to integrate into the clinic workflow through our EHR, or on a separate platform where we're going to be able to take all of the information a single patient has.

Make sure that we understand their risk factors, and come up with a graphical representation of this individual's natural history, and deploy our treatments accordingly. We've decided we have the right patient. Now, the question is the right therapy, and this is where the excitement around understanding biology comes into play.

If we're able to understand, most importantly, at the time of diagnosis that there are individual patients who respond preferentially to one treatment or another, imagine that precision biology approach that we'll be able to take away from this.

"I'm going to use this sequence of therapies because this is what I'm used to, or this therapy because I used it in another patient that had a similar phenotype." No. We're going to erase all that historical experience.

We're going to go to the tissue and decide that this patient, for example, would do best with anti-TNF first-line, this patient would be best with an IL-12 or an IL-23 blockade. This patient, integrin base, this patient, small molecule, or maybe this patient's combination.

If we understood the biology at the time of diagnosis, we will have a more precise way of addressing the biology, and we do believe that the first treatment is always the most effective.

If we can get that right, and we can understand the biology and what are some of the gene expression markers, or some of the more recent data on patients who may or may respond or not respond quickly to some of our anti-TNFs, for example, we're going to start to get a good idea of which patients we should be focusing on TNF versus non-TNF strategies.

I think that's a great place to start. I also share the fact that it's not just about efficacy and biology, it's also about safety. Pharmacogenetics and genomics, I've been studying this since starting to look at thiopurine metabolites back in the mid to late '90s, and the role of thiopurine methyltransferase.

That still remains an important part of our safety use of thiopurines, but now, we have a new genetic marker being implicated in the use of anti-TNFs, for example, with the HLA-DQA105 variant that's been shown from the PAD study in the UK to predict which patients will develop antidrug antibodies.

With the idea to use this to help us decide who do we use combination therapy on when we're doing anti-TNF versus patients we don't need to use combination therapy with an immunomodulator.

If we can imagine that just like decision-support tools on prognosis, that we're going to be able to integrate genetic and personalized biology info into some, again, decision-support tool that you can personalize and stratify patients into which their predominant biology is, then we're going to be able to be more precise with our treatment options.

Then, I ended with a snippet of the future, and what does the future look like? Is in order for us to be precise about the biology and understand risk in established IBD patients and develop the most predictive models, could it be that some of these models could be applied to what we call the preclinical—meaning, before patients become symptomatic in this pre-over-clinical symptoms -- and what we call secondary prevention -- and be able to build even new tools that may predict who's at most risk of getting disease?

Knowing the biology of those patients, and intervening and applying precision biology, too, imagine that idea of being able to even prevent clinical IBD. I think the future is bright. There's a journey of finding the right patient and finding the right therapy, and then precision can lead to prevention. Thank you so much.