Metabolic tumor volume (MTV) was reported at the 2022 annual meeting of the American Society of Hematology to be a predictive biomarker for patients with relapsed or refractory diffuse large B-cell lymphoma (rel/ref DLBCL) who were being treated with loncastuximab tesirine in the LOTIS-2 trial.1 Therefore, this biomarker, derived from PET/CT imaging, may guide individualized treatment by identifying patients who are most likely to respond well to loncastuximab tesirine. Physician’s Weekly spoke with Dr. Juan Pablo Alderuccio (University of Miami), lead author of the study, to reflect on the outcomes of the current post-hoc analysis.
Loncastuximab tesirine is a humanized antibody-drug conjugate (ADC) monoclonal antibody targeting the protein CD19 used to treat adults with large B-cell lymphoma. Although MTV is an established, strong prognostic factor in DLBCL, it is not yet known whether it may predict response to loncastuximab tesirine. Since predictive biomarkers are important to guide individualized treatment and optimize outcomes, Dr. Alderuccio and colleagues evaluated the association between quantitative PET/CT data and treatment response to loncastuximab tesirine in patients with rel/ref DLBCL who were enrolled in the LOTIS-2 trial (N=118). The primary objective was to assess the predictive value of MTV and total lesion glycolysis (TLG) on complete metabolic response (CMR).
The continuous variable log2 (MTV) was predictive of failure to reach a CMR (OR, 1.52; 95% CI, 1.22-1.89; P=0.002; AUC, 0.744). Similarly, one unit increase in log2TLG resulted in a decreased likelihood of achieving a CMR (OR, 1.49; 95% CI, 1.23-1.80; P<0.001; AUC, 0.758). Furthermore, increments on log2 (MTV) and log2 (TLG) were predictive of shorter event-free survival times (HR, 1.28; P<0.001; HR, 1.21; P<0.001) and shorter overall survival duration (HR, 1.38; P<0.001; HR, 1.29; P<0.001), respectively.
Physician’s Weekly interviewed Dr. Alderuccio to discuss the impact of these results.
PW: What is the added value of measuring metabolic tumor volume?
Dr. Alderuccio: Currently, there is no predictive biomarker to assess the response to different therapies in diffuse large B-cell lymphoma. We have been using the International Prognostic Index (IPI) score based on clinical variables since the 90s, with subsequent changes trying to improve the predicted value NCCN IPI score coming out in 2014.2,3
The problem we face in the clinic is that we do not base our treatment decision on the IPI score. There remains this unmet need to identify a biomarker to inform our choices. We are attempting to show that MTV might fill that role, here specifically for one treatment: loncastuximab tesirine. MTV has been shown to be a robust prognosis biomarker across different lymphomas in Hodgkin lymphoma4 and in diffuse large B-cell lymphoma.5
MTV is essentially a measurement of metabolically active tumor burden. In this specific study, we use the threshold of 41%, and all the lesions that are about that threshold are included in the final tumor volume. The initial data set we used was derived from the two trials that led to the approval of loncastuximab tesirine in patients with two or more lines of therapy in diffuse large B-cell lymphoma.6,7
We presented a preliminary analysis, and we are now finalizing the total clinical trial cohort. But the initial data we presented were definitely interesting. We try to assess the predictive value of MTV. We want to predict which patients will achieve complete response on this therapy.
Unfortunately, in the relapsed/refractory setting of DLBCL, we know that most patients who do not achieve a complete response have poor outcomes. Secondly, we assess the prognosis implications in progression-free survival and overall survival based on a screening on pre-treatment metabolic tumor volume. We also assess the tumor lesion glycolysis, and other parameter metrics, to try to compare how they perform with MTV.
We delineated a specific cut-off point that predicts outcome in this initial data set of 118 patients. We found, albeit in this small dataset of only 118 patients, that MTV is a robust predictor of response to loncastuximab tesirine. Our hypothesis warrants further investigation.
Will MTV be developed into a useful biomarker that can be clinically actionable?
Yes, assuming we are able to externally validate our findings. I think the next step for this particular study is to complete the clinical trial cohort and see what the cut-off points are that we identify as a predictor of complete metabolic response and survival in this DLBCL population treated with loncastuximab tesirine. We will perform internal validation with statistical bootstrapping methodology because it is not a very large clinical dataset. Once we have determined those cut-off points, the next step will be to externally validate the prognostic value of MTV in other clinical trials treating patients with DLBCL with loncastuximab tesirine. Assuming that is successful and external validation holds up, I think we will be able to predict what patients will benefit most from this approach, and I think the next step will be to test this biomarker prospectively in clinical trials.
Will MTV enable risk stratification to determine specifically which patients will benefit most from loncastuximab tesirine?
Exactly, because currently in lymphoma, we don’t have any biomarker that is able to predict outcomes. This will be highly important for the lymphoma community.
Is there a correlation between genomics and MTV?
We haven’t yet explored the correlation between genomics and higher MTV. We are, however, planning that very analysis as a priority going forward.
But if I wee to speculate, I think we can expect that patients with higher MTVs will be characterized by more complex genomic features associated with poor prognosis, such as TP53 mutation. The challenge with diffuse large B-cell lymphoma is that there are no specific mutations that are commonly associated with worse prognosis like in other types of diseases; it is a very heterogeneous disease. Nevertheless, I think that there can be a correlation there between genomics and MTV. An additional step forward will be to correlate MTW with radiomics and eventually even with radiogenomics. The challenge there will be to find specific features extracted from imaging data that may indicate the presence of specific genetic mutations.
I think it is very exciting times for lymphoma research. I think the data extracted from imaging data can be extremely useful in treatment prediction for outcomes and survival. Moreover, the incorporation of machine learning is allowing us to process very large data sets and then go back to our clinic and match what we find in the patient with what we find in their tumor profiles, genomically, metabolically, or radiographically.
We need further validation, especially external validation, as any biomarker does, in order to prove the efficacy to discriminate between respondents and non-respondents and prognosis. Then, I think the next step will be to prove this biomarker prospectively. Hopefully, we will be able to incorporate MTV into our clinical practice in the near future.