Evaluating A-HIPI vs Traditional Scores in Advanced Hodgkin Lymphoma: Andrew Evens, DO, MBA, MSc

Continuous variable modeling such as the Hodgkin lymphoma international prognostic index (A-HIPI) improves prognostic accuracy compared with the traditional International Prognostic Score (IPS), says Andrew Evens, DO, MBA, MSc, deputy director for clinical services and chief physician officer, Rutgers Cancer Institute.

This transcript was lightly edited; captions were auto-generated.

Transcript

Can you provide an overview of this study on the clinical prognostication in the SWOG 1826 trial?

This is an oral presentation coming up at this year's 2025 ASH meeting entitled "Clinical prognostication in the SWOG 1826 randomized clinical trial of nivolumab-AVD vs brentuximab-AVD,' looking at the performance of advanced-stage Hodgkin lymphoma international prognostic index, otherwise known as the A-HIPI. It is an important analysis because, number 1, we know that Hodgkin lymphoma is very treatable and curable, but there certainly is still a subset of patients who experience treatment resistance and potentially death. Number 2, there is a desire to really have accurate, not just prognostic models, but predictive modeling, to understand who does really, really well with current treatment and who does not as well and could potentially benefit from newer or different treatment. And that's what modeling, whether prognostic or predictive, can help you do. And so we applied the A-HIPI to this very contemporary, recently published study from S1826 that was recently published in The New England Journal of Medicine earlier this year, I should say.

How does the A-HIPI improve prognostic accuracy for advanced-stage Hodgkin lymphoma compared with the traditional IPS?

The traditional prognostication was known as the International Prognostic Score that was published in the 1990s and was a very important analysis and had really been the prognostic score for advanced-stage Hodgkin lymphoma for the ensuing 2025 years. There had been a few publications in the more modern era that said it may be overpredicting and not as accurate as we thought, so through a consortium called HOLISTIC, which we established in 2018, HOLISTIC stands for Hodgkin lymphoma, international study for individual care. It is a consortium where we have now amalgamated over 30,000 Hodgkin lymphoma cases into a centralized database with a common data model, and it constitutes about two-thirds of patients' individual, discrete cases from large randomized clinical trials over the last 15 years, as well as about a third of the cases of the 30,000 from large, mostly prospective registries, and those are important because they're enriched with longer-term follow-up. There was an initial publication of a HOLISTIC and advanced-stage Hodgkin lymphoma called the ALPPI that was published in The Journal of Clinical Oncology in 2003, and it was compared to that aforementioned IPS, and it was shown to have superior discrimination, which is separating into low- and high-risk groups, as well as especially superior calibration. Calibration is another way to say more absolute risk estimates of a model, or another way to say it for calibration, which is really important; it informs clinicians how the predicted outcome relates to the actual outcome in external groups of patients. And so we published that there's an online tool. It's been since validated in four other ensuing publications, from an Italian set, from a Nordic data set, a Brazilian data set, etc., but therapy changes in a good way in many cancers, and in this case, Hodgkin lymphoma, off of the heels of the S1826 study, showed that nivolumab-AVD was superior to brentuximab vedotin-AVD. You never want to assume that a model stays relevant through different treatments. You need to analyze it and test it, and that's what we did. We assessed the prognostic performance, discrimination, and calibration of the A-HIPI in this very contemporary data set.