AUTHORS
Jacqueline F. Wang, MD, Manik Uppal, MD2, Venkatraman Seshan, PhD3, Michelle Okwali, MPH4, Xinjun Wang, PhD5, Suheil Albert Atallah-Yunes, MD6, Kevin David, MD7, Zachary Epstein-Peterson, MD8, Lorenzo Falchi, MD9, Paola Ghione, MD, MS10, Steven Horwitz, M.D.11, Anita Kumar, MD12, Efrat Luttwak, Luttwak, MD13, Alison Moskowitz, MD14, Colette Owens, MD15, Maria Lia Palomba, MD16, Raphael Steiner, MD17, Robert Stuver, MD18, Pallawi Torka, MD19, Jasmine Zain, M.D.20, Andrew Intlekofer, MD, DPhil21, Santosha Vardhana, MD, PhD22, Quinlan Sievers, MD, PhD23, Andrew Zelenetz, MD24, Ariela Noy, MD25, Mark Roschewski, MD26, Gilles Salles I, MD, PhD27, Jennifer Kimberly Lue, M.D.28
AFFILIATIONS
1 Memorial Sloan Kettering Cancer Center, New York, New York, United States
2 Icahn School of Medicine at Mount Sinai, New York, New York, United States
3 Memorial Sloan Kettering Cancer Center, New York, New York, United States
4 Memorial Sloan Kettering Cancer Center, New York, New York, United States
5 Memorial Sloan Kettering Cancer Center, New York, New York, United States
6 Memorial Sloan Kettering Cancer Center, New York, New York, United States
7 Memorial Sloan Kettering Cancer Center, New York, New York, United States
8 Memorial Sloan Kettering Cancer Center, New York, New York, United States
9 Memorial Sloan Kettering Cancer Center, New York, New York, United States
10 Memorial Sloan Kettering Cancer Center, New York, New York, United States
11 Memorial Sloan Kettering Cancer Center, New York, New York, United States
12 Memorial Sloan Kettering Cancer Center, New York, New York, United States
13 Memorial Sloan Kettering Cancer Center, New York, New York, United States
14 Memorial Sloan Kettering Cancer Center, New York, New York, United States
15 Memorial Sloan Kettering Cancer Center, New York, New York, United States
16 Memorial Sloan Kettering Cancer Center, New York, New York, United States
17 Memorial Sloan Kettering Cancer Center, New York, New York, United States
18 Memorial Sloan Kettering Cancer Center, New York, New York, United States
19 Memorial Sloan Kettering Cancer Center, New York, New York, United States
20 Memorial Sloan Kettering Cancer Center, New York, New York, United States
21 Memorial Sloan Kettering Cancer Center, New York, New York, United States
22 Memorial Sloan Kettering Cancer Center, New York, New York, United States
23 Memorial Sloan Kettering Cancer Center, New York, New York, United States
24 Memorial Sloan Kettering Cancer Center, New York, New York, United States
25 Memorial Sloan Kettering Cancer Center, New York, New York, United States
26 Memorial Sloan Kettering Cancer Center, New York, New York, United States
27 Memorial Sloan Kettering Cancer Center, New York, New York, United States
28 Memorial Sloan Kettering Cancer Center, New York, New York, United States
ABSTRACT
Introduction: Central nervous system (CNS) relapse occurs in 5-15% of patients (pts) with diffuse large B-cell lymphoma (DLBCL) and carries a dismal prognosis (Eyre et al., Lancet Oncology, 2022). Risk models such as CNS-IPI rely solely on clinical factors, omitting genetic characteristics (Schmitz et al., JCO, 2016). Although rare, other B-cell lymphomas can involve the CNS. The current gold standard for CNS involvement is CSF flow cytometry/cytology and/or brain biopsy. However, sensitivity of flow cytometry is suboptimal. We hypothesized that evaluation of CSF using next-generation sequencing (NGS) may improve detection of CNS disease (Grommes et al., Blood, 2019). We conducted a retrospective analysis of pts with B-cell lymphoma who underwent CSF-based targeted NGS using MSK IMPACT-Heme at Memorial Sloan Kettering.
Methods: Pts with B-cell lymphoma from 2015 to 2025 with paired CSF flow cytometry/IMPACT-Heme were included. CSF analysis was obtained at initial diagnosis or relapse. Time zero (T0) was the date of paired CSF flow cytometry and IMPACT-Heme. CNS involvement was defined as positive (pos) CSF flow, cytology and/or brain biopsy. PFS and OS from T0 were estimated by the Kaplan-Meier method. Mutations were visualized using ComplexHeatmap R package.
Results: Among 111 pts with CSF IMPACT-Heme, 75 had paired CSF flow/IMPACT. This cohort was 51% male with median age of 66 years (range 38-86) and a median of 3 lines of therapy. Thirty-seven pts (49%) had baseline CNS involvement with histologies including PCNSL (n=27), DLBCL with synchronous SCNSL (n=6), Bing-Neel Syndrome (WM-BN; n=2), HGBCL (n=1) and ocular lymphoma (OL; n=1). Thirty-eight pts (51%) did not have baseline CNS involvement, including DLBCL (n=20), OL (n=5), HGBCL (n=4), WM-BN (n=2), MCL (n=3) and other (n=4). Among pts with CNS involvement at baseline or relapse (n=56), 91% (n=51) received methotrexate-based therapy. Thirteen pts (17%) underwent ASCT; 8 had CNS involvement and received ASCT as consolidation. Nine pts (12%) received CAR T-cell therapy at relapse.
Of the 24 pts without baseline CNS involvement (DLBCL, n=20; HGBCL, n=4), 14 (DLBCL, n=13; HGBCL, n=1) later had CNS relapse, with low-risk CNS-IPI scores (0-1) in 2, intermediate-risk (2-3) in 6, and high-risk (4-6) in 6 pts. Of these 14 pts, non-GC subtype was observed in 50% (n=7). In contrast, 43% of pts (n=3) with synchronous SCNSL (DLBCL, n=6; HGBCL, n=1) were non-GC subtype. The remaining 10 pts without CNS involvement (DLBCL, n=7; HGBCL, n=3) were mostly (80%) GC subtype. Of the 20 pts with DLBCL without baseline CNS involvement, 10 received CNS prophylaxis (ppx), 3 of whom had CNS relapse despite ppx. All 4 HGBCL pts without baseline CNS involvement received CNS ppx, 1 of whom had CNS relapse. CNS ppx included IV and/or intrathecal methotrexate.
Forty-one pts (41/75, 55%) had CSF IMPACT positivity: 25 with concordant CSF flow pos/IMPACT pos (double pos) status and 16 with discordant CSF flow negative (neg)/IMPACT pos disease. The concordant group included DLBCL with CNS relapse (n=7), PCNSL (n=5), SCNSL (n=4) and WM-BN (n=4). The discordant group included PCNSL (n=7), OL (n=2), SCNSL (n=2), HGBCL without CNS involvement (n=2), DLBCL without CNS involvement (n=1), other (n=2). In the discordant group, all pts with HGBCL and DLBCL without CNS involvement received CNS ppx; none had CNS relapse. Twenty-seven pts (36%) had double neg status; 7 pts were CSF flow pos/IMPACT neg (9%).
At a median follow-up of 2.9 years, 1-year PFS was 74%, 74%, 71%, 47% for double neg, CSF flow neg/IMPACT pos, CSF flow pos/IMPACT neg, and double pos, respectively (p=0.02). The 1-year OS was 89%, 94%, 71% and 55% for the same respective groups (p=0.02). On univariate analysis, CSF flow positivity (p=0.004) and IMPACT positivity (p=0.04) were associated with inferior OS. The most common genomic alterations in the IMPACT pos group were in MYD88, KMT2D, PIM1, ARID1A and BTG1.
Conclusions: Concordant CSF flow pos/IMPACT pos status was associated with worse PFS and OS. Pts with discordant CSF flow neg/IMPACT pos status had outcomes comparable to double neg pts, though the small, heterogeneous cohort limits definitive conclusions. Genomic results were consistent with prior studies, with an enrichment of MYD88 L265P. Our study highlights the need for prospective trials to evaluate CSF NGS as a companion to CSF flow; an ongoing study evaluating CSF MSK IMPACT-Heme as a complementary tool is underway (NCT06736613).
DISCLOSURES
Jacqueline F. Wang, MD: No relevant conflicts of interest to declare.