Publication

Article

Evidence-Based Oncology

January 2024
Volume30
Issue 1
Pages: SP79-SP80

Data Reveal Racial, SDOH Factors Linked to OS Differences in CLL

It’s been nearly a decade since the first Bruton tyrosine kinase (BTK) inhibitor for chronic lymphocytic leukemia (CLL), ibrutinib, reached the market and changed the treatment landscape for patients with this disease. But are all patient groups benefiting? That was a question that investigators in The US Oncology Network wanted to explore as they used the network’s database to explore whether survival outcomes showed differences by race or social determinants of health (SDOH).

Results were presented in an abstract December 9, 2023, during the 65th American Society of Hematology (ASH) Annual Meeting and Exposition in San Diego, California.1

Ira Zackon, MD

Ira Zackon, MD

Ira Zackon, MD, a hematologist/oncologist with New York Oncology Hematology, is the senior medical director with Ontada, a data science company that is part of McKesson. Zackon, senior author on the abstract presented at ASH, spoke with Evidence-Based Oncology (EBO) about the findings. The following has been edited lightly for clarity.

EBO: What were the key takeaways from the review of the data on CLL?

Zackon: This was sourced in The US Oncology Network database, using the electronic health record code. We were looking at a large population of US CLL patients at the community level, with a particular focus on any differences in the patient characteristics within the CLL population—and importantly, on the outcome of overall survival.

We focused on race and any socioeconomic factors. This included over 12,000 CLL patients, and we captured race for almost 90%, so it’s a good representation. First, we found in the population, 91% of the patients were White and 5% were Black, and the 4% were another race. This generally in line with the known data of the current state of CLL; there is lower incidence in the Black population.

Importantly, when we looked at any differences in the patient characteristics, we found that Black patients had a younger mean age by about 2 years, aged 68 years vs 70 years, and there was no difference in the male to female ratio. But Black patients did present with more advanced stage disease—stage III or IV CLL—at the time of diagnosis.
We looked at socioeconomic factors, or total household income and levels of education. This was from what we call structured data; the physician is giving the care and entering in the data, but some are in structured fields where we can easily extract that data for the purposes of research and analysis.

There were clear differences for patients with a lower household income, defined as less than $30,000 annually vs more than $30,000, and in the levels of education, whether completing just high school, college, or postgraduate studies. Importantly, when looking at the outcome of overall survival, we measured from the time of their first visit, not their time of diagnosis, so we can capture the journey within their care at the time they enter into the database. There was a very significant difference in overall survival.

The patients selected were treated from 2015 through mid-2023, so it’s an 8-year period of data. And that was to reflect the period of transformation that we’ve seen in CLL, with the introduction of BTK inhibitors in 2014, and now 3 in-class agents in the CLL space, and then subsequently, BCL2 inhibitors. It’s really changed the way we treat CLL, and it’s changed the outcomes in CLL. We wanted to capture that era, so that’s the time frame [we selected] for the patients.

We found that when looking at mortality, and we know internally in our dataset, we have a validated high capture of mortality within the EHR [electronic health record] and we connect to outside data sources of mortality. In the Black population, there was a 61% rate of death within this 8-year data period, whereas in the White population it was only 43%, so there was almost a 20% difference in overall survival.

Then, when we looked at the different patient characteristics that we used in this study, if you adjusted for the socioeconomic factors, the difference in race did not reach statistical significance any longer. But household income remained statistically significant; it remained a potential driver of differences.

Now, this is at a very high level; we’re using structured data only. But I think the important conclusions from the study are that we’re seeing what looks like a significant difference in overall survival. And that may relate to both to race and the associated lower socioeconomic factors.

EBO: What data need more examination?

To explore that next level of understanding, I should say that those socioeconomic factors—educational level and household income—we only captured unstructured data in about 25% of the patients, so there’s a significant missingness in that data. We see what we call the wider confidence intervals. And yet, it still retained a statistically significant difference.

I think to get at that next level of understanding, you would need to do chart reviews—chart abstraction—to examine we call the unstructured data. These are the narrative portions of the electronic health record, the attached documents. We have to look at both to be sure that there’s no differences in our current understanding of CLL biology, [such as] TP53 mutation status.… We’d have to look at actual treatments delivered and the duration of treatment that patients receive, and then get more granular into social determinants of health.

Again, the background for this study is that we understand the improvement in outcomes with our treatments based on prospective trials, but we certainly have increasingly recognized what we would call social determinants of health. So outside of disease biology, outside of treatments, what are the factors that may impact ultimately an outcome? We would need to look at comorbidities and health and then have more granular breakdown.

EBO: How will you measure SDOH going forward?

Zackon: What we’ve introduced now, and it’s going forward into our EHR and then into the database, is what’s called the Distress Thermometer. If you’re familiar with that, it’s quite comprehensive and capturing many facets of our life.2 So it does capture financial information; there are also aspects on social support, transportation issues, emotional and mental health, spiritual health. A variety of factors that could impact outcomes will be entered into the data, and we will be able to harvest that type of analytics to gain more meaningful understanding. That’s what ultimately drives what can we do to make sure that all our CLL patients in the United States are receiving the full potential benefits of the therapies that we have now and in the future. 

References
1. Su Z, O’Sullivan AK, Marcus A, Zackon IL. Racial and socioeconomic disparities among patients with chronic lymphocytic leukemia (CLL) treated in the US community oncology setting. Presented at: 65th American Society of Hematology Annual Meeting and Exposition, December 9-12, 2023; San Diego, CA. Abstract 2406. doi:10.1182/blood-2023-182498

2. NCCN. Clinical Practice Guidelines, Distress management, version 1.2024. Accessed December 19, 2023. https://www.nccn.org/docs/default-source/patient-resources/nccn_distress_thermometer.pdf

Related Videos
Dr Cesar Davila-Chapa
Screenshot of an interview with Nadine Barrett, PhD
Milind Desai, MD
Masanori Aikawa, MD
Neil Goldfarb, GPBCH
Mabel Mardones, MD.
Mei Wei, MD, an oncologist specializing in breast cancer at Huntsman Cancer Institute at the University of Utah.
Alexander Mathioudakis, MD, PhD, clinical lecturer in respiratory medicine at The University of Manchester
Mabel Mardones, MD.
Screenshot of Susan Wescott, RPh, MBA
Related Content
AJMC Managed Markets Network Logo
CH LogoCenter for Biosimilars Logo