A newly devised, validated prognostic tool – the Burkitt Lymphoma International Prognostic Index – can consistently identify low-risk patients who might benefit from treatment de-escalation, and high-risk patients who are unlikely to be cured with current therapies and may require novel approaches, investigators said.
In a cohort of patients treated at international sites, patients with a low-risk score on the BL-IPI had a 3-year progression-free survival (PFS) rate of 96%, and 3-year overall survival rate (OS) of 99%. In contrast, the 3-year PFS rate for patients in the high-risk category was 63%, and the 3-year OS rate was 64%, reported Adam J Olszewski, MD, from the Lifespan Cancer Institute at Rhode Island Hospital and The Miriam Hospital, both in Providence.
“The Burkitt Lymphoma International Prognostic Index – or the ‘BLI-PI’ [‘blippy’] as it was inevitably called – is a novel prognostic index that is specific to Burkitt lymphoma. It has been validated with sufficient calibration and discrimination in external data sets to allow for simple stratification and comparison of risk distribution in geographically diverse cohorts,” he said in an oral abstract presented virtually during the annual meeting of the American Society of Hematology.
Inconsistent criteria
There is a need for a Burkitt-specific index, he said, because of significant differences in age, stage at presentation, and abnormal lactate dehydrogenase (LDH) levels between patients with Burkitt and those with diffuse large B-cell lymphoma (DLBCL), and because historical definitions of “low-risk” Burkitt lymphoma have been inconsistent, with less than 10% of patients falling into this group, leaving the remainder in a undifferentiated “high-risk” category.
“Burkitt lymphoma is considered highly curable, but current therapy requires administration of dose-intense chemoimmunotherapy for which there are many chemotherapy backbone regimens developed across the world, and used mostly locally. These are often studied in phase 2 studies with limited sample sizes, which makes it difficult to compare populations across trials,” Dr. Olszewski said.
A validated prognostic index can help clinicians and researchers compare cohorts and can be used to help design future trials, he added.
To devise the BL-IPI, the investigators first selected a retrospective cohort of 570 adults with Burkitt lymphoma treated at 30 U.S. centers for whom data on outcomes were available.
They determined the best prognostic cutoffs for age, LDH, hemoglobin and albumin levels, and identified independent risk factors using stepwise selection in Cox regression and lasso regression analysis, a machine learning approach. The variables included age; sex; HIV-positivity status; loss of MYC rearrangement; performance status; stage; nodal involvement; marrow involvement; central nervous system involvement; and LDH, hemoglobin, and albumin levels.
For validation, they pooled data from European, Canadian, Australian, and U.K. studies to identify 457 patients for whom retrospective treatment and outcomes data were available.
The derivation and validation cohorts were similar in most respects, expect for a higher proportion of patients with Eastern Cooperative Oncology Group performance status scores of 2 or higher in the validation cohort (22% vs. 35%), and a higher proportion of patients with CNS involvement in the U.S.-based derivation cohort (19% vs. 10%, respectively).
Therapy also differed markedly between the U.S. and international cohorts, with about 30% each of U.S. patients receiving either the CODOX-M/IVAC (cyclophosphamide, vincristine, doxorubicin, high-dose methotrexate/ifosfamide, etoposide, and high-dose cytarabine) regimen, DA-EPOCH-R (dose-adjusted etoposide, prednisone, vincristine, cyclophosphamide, doxorubicin, and rituximab) regimen, or hCVAD/MA (fractionated cyclophosphamide, vincristine, doxorubicin, and dexamethasone alternating with high-dose methotrexate and cytarabine) regimen, and the remaining 10% receiving other, unspecified therapy.
In contrast, 65% of the patients in the international (validation) cohort received CODOX-M/IVAC, 10% and 9%, respectively, received DA-EPOCH-R and hCVAD/MA, and 16% receiving other regimens.
Rituximab was administered to 91% of U.S. and 95% of international patients.