Optimizing Treatment Alternatives for Relapsed/Refractory T Cell Lymphomas
In This Article
- Massachusetts General Hospital leads an international consortium of clinicians and researchers focused on improving treatment options and outcomes for T cell lymphomas
- Retrospective data from a large global cohort enabled the development of a prognostic model that facilitates clinical decision-making for patient care
- Study results suggest that single-agent drugs may demonstrate increased therapeutic efficacy relative to conventional chemotherapy regimens
Physician-scientists at Massachusetts General Hospital recently published research findings from the peripheral T-cell lymphoma (PETAL) consortium. Using over 10 years of global data from patients with mature T/natural killer cell lymphoma (MTCL/MNKCL), researchers created a prognostic model to optimize second-line treatments for relapsed/refractory (R/R) patients.
Subscribe to the latest updates from Cancer Center Advances in Motion
"Access to data from a diverse patient population offered a unique opportunity to assess outcomes to different therapeutic strategies on a global scale," says Salvia Jain, MD, Mass General clinician-investigator, assistant professor of medicine at Harvard Medical School, and principal investigator and founding member of the PETAL consortium. "The result is a prognostic index enabling clinicians to refine treatment options for the sickest patients."
Tackling a Rare Cancer With Few Treatment Options
Although T cell lymphomas (TCLs) account for about 7% of non-Hodgkin lymphomas diagnosed in the United States, their frequency is highly heterogeneous across different ethnic and demographic populations. Among over 30 TCL subtypes, MTCL and MNKCL are extremely rare (less than 2% of TCL cases) and highly aggressive.
First- and second-line treatments for TCLs generally involve chemotherapy regimens that may be combined with radiation, depending on the subtype. For eligible patients, autologous or allogeneic hematopoietic stem cell therapy (HSCT) can be administered to achieve further remission or address relapse or acquired therapeutic resistance. Regardless of the approach, up to 40% of cases become refractory to standard treatments, including multi-agent chemotherapy regimens.
Dr. Jain explains that the difficulty lies in determining the best approach to treating an extremely rare, highly aggressive disease with multiple subtypes across diverse patient populations.
"For patients that achieve some level of remission, up to 60% relapse," she explains. "Ultimately, around 70% of patients present relapsed or refractory disease, and the majority die within two years."
Rare diseases present multiple roadblocks to identifying novel and effective therapeutic options. In the case of MTCL/MNKCL, Dr. Jain highlights two:
- Biological characterization: Understanding these cancers requires tumor data from multiple institutions and a collective focus on identifying and leveraging vulnerabilities.
- Targeted drug development: The steps required to develop and evaluate new drugs are extremely time- and cost-intensive, making the targeting of rare diseases difficult to justify.
This also presents a major hurdle for clinical trials. Limited patient populations hinder an ability to obtain statistical power in outcomes or identify specific patient groups possibly benefitting from certain therapies. For rare TCLs, trials populated with patients presenting significant disease heterogeneity may be biased against targeted treatments that could show promise for small subsets of patients.
"Anyone who has worked in this field for an extended period understands the challenges," says Dr. Jain. "As a young investigator, I realized that addressing the poor survival outcomes in these patients would require a global community of like-minded colleagues willing to share resources and find solutions."
Figure 1
Paving the way for breakthroughs by creating an infrastructure for future clinical trials to test novel treatments.
Predicting Patient Prognosis to Streamline Treatment Plans
The consortium's ability to generate enormous volumes of high-dimensional data offered access to randomized datasets involving global patient populations. Capitalizing on machine learning methods offered an unprecedented opportunity to retrospectively assess treatment outcomes across ranges of clinical characteristics, TCL subtypes, and therapeutic approaches.
The consortium's first initiative aimed to identify risk factors and subgroups within R/R MTCL/MNKCL that lead to variability in treatment outcomes and overall survival (OS). Because these patients have few second-line treatment options, the goal was to develop a way to stratify the most difficult-to-treat patients according to risk (as defined by a worse-predicted OS).
"Previous predictive models focused on prognostication of newly diagnosed MTCL/MNKCL patients," Dr. Jain explains. Given the variable outcomes observed in R/R patients with an already poor prognosis, this met the need for a prognostic model supporting a personalized approach to their treatment.
The authors evaluated 763 R/R MTCL/MNKCL patients receiving second-line treatment from 13 medical institutions in 10 countries. Their analysis identified six independent predictors of OS, including age, refractory disease status, histological subtype, and molecular signature.
Incorporating these into a prognostic index for R/R TCL (PIRT) score allowed the estimation of OS from the start of second-line treatment. The score ranges from 0 to 6, with a score of 1 assigned to each of the six predictors. R/R patients can then be stratified into low- (score: 0–1), intermediate- (score: 2–3), and high-risk (score: 3–4) groups. Dr. Jain advises that the PIRT score represents a first step in helping to facilitate clinical decisions regarding treatment approaches for different patients.
Because a higher PIRT score correlates with worse OS, treatment strategies can be tailored accordingly. For example, a low-risk PIRT score suggests a patient may continue responding to a standard therapy. By contrast, intermediate- or high-risk scores might warrant a shift away from chemotherapy toward HSCT or enrollment in specific clinical trials. This also facilitates populating trials with patients matching a similar clinical profile and with a higher likelihood of responding to non-standard approaches.
The PIRT Score Calculator is currently available for use by clinicians on the PETAL consortium website.
Figure 2
PETAL: A Global T-Cell Lymphoma Consortium integrating machine learning and genomics to drive discovery and predict outcomes for newly diagnosed and relapsed/refractory T-cell and NK-cell lymphomas.
Expanding Utilization of Novel Therapies
Despite the development of multiple classes of single-agent (SA) drugs, streamlining how and when they should be used has also been difficult. These drugs include immunomodulatory agents, epigenetic therapies, and small-molecule inhibitors, all of which demonstrate varying degrees of success according to the timing of administration in different patients.
However, these drugs are not available worldwide, and studies reporting their efficacy are limited in size and location. Dr. Jain notes that rigorous comparisons of novel therapies against standard approaches normally require phase III clinical trials with large patient populations, which are nearly impossible for rare diseases.
"It's difficult to champion the use of these drugs in any setting without solid evidence of their safety and effectiveness," she explains.
To address this, the authors used data from the same global cohort of R/R MTCL/MNKCL patients to compare the efficacy of SA drugs versus chemotherapy delivered as second-line treatment. The analysis specifically measured treatment response from the beginning of second-line treatment to the start of third-line treatment, death, or loss of follow-up. Their results revealed a significant improvement in progression-free survival in patients receiving SA drugs regardless of cancer subtype.
"Previous studies indicated that some SA drugs targeting a specific disease pathway showed significant promise as alternative treatment options," Dr. Jain explains. "This evidence generated from real-world data supports their potential use earlier and more frequently during the course of the disease."
Figure 3
Meet the PETAL team (from left to right): Sean McCabe, Kusha Chopra, Ronald Nemec, Jamie Weller, Olivia Economides, MPH, Anna Rider, Shambhavi Singh, MD, PhD, Khyati Kariya, PhD, Khyati Kariya, PhD, and Salvia Jain, MD.
Transforming the Future of Lymphoma Treatment
The PETAL consortium's second initiative is currently underway and aims to reveal molecular targets that will enable personalized approaches to treatment across TCL subtypes. The consortium's success thus far has increased its momentum in acquiring new collaborations, including interest from pharmaceutical companies and numerous lymphoma centers of excellence worldwide.
"We currently have 10 additional sites on track to begin enrolling patients in a longitudinal study of newly diagnosed R/R TCL patients," says Dr. Jain. This includes the establishment of biobanks at each site and eventual collaborations in analyzing the collected samples. She notes that Mass General has already enrolled 35 patients and collected thousands of samples in the lymphoma biobank at the Jon and Jo Ann Hagler Center for Lymphoma.
"Our intention is to encourage pharmaceutical companies and academic researchers to refine how clinical trials are designed for these diseases," she says. "Ultimately, the goal is to increase patient access to novel therapies."
Regarding the ambition of these goals, Dr. Jain emphasizes that Mass General's ecosystem is designed to support initiatives on the scale of establishing and managing worldwide collaborations.
"When I explained the concept of the consortium, I received immediate encouragement and the support required to make it happen," she says. "Their approach is to facilitate access to the resources necessary to optimize patient care regardless of the scope of the project."