Key Points
-
•
EMD was seen at a higher incidence in patients with R/R CRLF2r B-ALL than in those without CRLF2r.
-
•
Increased surveillance for non-CNS EMD may be warranted in patients at higher risk for EMD.
Visual Abstract
Abstract
Cytokine receptor–like factor 2 rearrangement (CRLF2r), an aberration causing a Philadelphia chromosome–like gene expression profile and a poor prognostic marker in B-cell acute lymphoblastic leukemia (B-ALL), endows relative chemotherapeutic resistance. Extramedullary disease (EMD), including both central nervous system (CNS) and non-CNS involvement, can also be associated with worse outcomes. The relationship between CRLF2r and EMD is unknown. Based on an initial observation of patients with relapsed/refractory (R/R) CRLF2r B-ALL and EMD, a single-center retrospective review was conducted to evaluate the incidence of EMD in patients with CRLF2r B-ALL. Across 177 children, adolescents, and young adults with R/R B-ALL, 21 (11.9%) harbored CRLF2r, of whom 19 (90.5%) had EMD at some point during their treatment course. In contrast, only 78 of the 156 patients (50%) without CRLF2r ever had EMD (P = .0003). In the CRLF2r cohort, 8 (38.1%) presented with EMD at diagnosis: 7 with CNS and 1 with non-CNS EMD. At relapse, 17 (80.9%) had EMD: 4 (19%) had CNS EMD, 9 (42.9%) had non-CNS EMD, and 4 (19%) had both. Isolated EMD was observed in 2 patients (11.8%) at first relapse and in 3 (17.6%) with multiple relapses. Given these findings, further study to evaluate this potential association between EMD and CRLF2r in R/R B-ALL is indicated.
Introduction
Among key genetic alterations in B-cell acute lymphoblastic leukemia (B-ALL), cytokine receptor–like factor 2 rearrangement (CRLF2r) has emerged as a significant prognostic marker because of its association with JAK2 pathway activation, chemotherapeutic resistance, and prevalence in high-risk B-ALL.1 Notably, CRLF2r represents a major subset of kinase-activating alterations that exhibit a gene expression profile similar to Philadelphia chromosome–positive (Ph+) ALL (ie, Ph like)1, 2, 3 and is associated with particularly poor prognosis.3
Another important factor in B-ALL is the presence of extramedullary disease (EMD).4,5 Central nervous system (CNS) involvement in newly diagnosed B-ALL has been identified in ∼3% to 10% of the cases using cytospin-based cytology.4 With the implementation of cerebrospinal fluid (CSF) flow cytometry, detection rates have increased, ranging from 17% to 41%.4 Although CNS infiltration is more frequently observed in patients with hyperleukocytosis and in T-cell ALL,5 specific cytogenetic and molecular alterations in B-ALL, such as t(1;19)(q23;p13) E2A::PBX1, BCR::ABL, and KMT2A rearrangements, are also linked to a higher risk of CNS involvement.5,6 Although routine CNS surveillance and prophylaxis are standard in B-ALL therapy, non-CNS EMD is associated with heterogeneous manifestations and likely underdiagnosed because of a lack of standardized screening.7 Accordingly, there is limited understanding of non-CNS EMD involvement and correlation with specific genetic/molecular subtypes of B-ALL.
Based on the observation of a series of patients at our institution with relapsed/refractory (R/R) CRLF2r B-ALL who had EMD, we sought to comprehensively analyze this cohort of patients.
Methods
Study population
This was a single-center retrospective analysis of 177 children, adolescents, and young adults with R/R B-ALL who were evaluated for 1 of the 4 institutional CD19- and/or CD22-targeted chimeric antigen receptor (CAR) T-cell clinical trials at the National Cancer Institute (NCI) between 2012 and 2025 (ClinicalTrials.gov identifiers: NCT01593696, NCT02315612, NCT03448393, and NCT05442515). Two patients were excluded from this analysis: one with Ph-like B-ALL with unconfirmed CRLF2r and another with a lack of available clinical data. All participants provided informed consent, and all samples were collected in accordance with the Declaration of Helsinki under National Institutes of Health Institutional Review Board–approved clinical protocols. This retrospective analysis was approved by the National Institutes of Health Institutional Review Board (ClinicalTrials.gov identifier: NCT03827343).
Disease characterization
EMD was defined by involvement of B-ALL outside the bone marrow (BM), established by tissue biopsy, CSF sampling, or imaging (eg, 8-fluorodeoxyglucose positron emission tomography/computed tomography [CT], CT, or magnetic resonance imaging). Sites of EMD disease were obtained via chart review and categorized as CNS EMD and/or non-CNS EMD (including testicular involvement), assessed from initial diagnosis through evaluation and/or treatment at the NCI. Because routine imaging is typically not performed in B-ALL, the rationale for imaging in patients was generally for any clinical suspicion of non-CNS EMD, prior history of known EMD, or active concern for non-CNS EMD, such as incidental imaging findings on scans performed for alternative reasons (eg, CT scan to evaluate for infection). Isolated extramedullary relapse was defined as relapse occurring in an extramedullary site without concurrent BM involvement.
Disease burden in the BM was classified using standard morphologic criteria: M1 <5%, M2 5% to 25%, and M3 >25%. Minimal residual disease (MRD) was evaluated by flow cytometry, with MRD positivity defined as ≥0.01% B-ALL among total mononuclear cells.
CSF was assessed using standard CSF assessment: CNS-1 (no blasts), CNS-2 (blasts with <5 total white blood cells per millimeter cube), and CNS-3 (blasts with ≥5 total white blood cells per millimeter cube or grossly traumatic). Patients with CNS-2 and CNS-3 status were categorized as having CNS EMD. Cytogenetics and molecular data for each patient were collected from available results in the medical record.
Newly diagnosed cohort
Given that our cohort included only patients with R/R disease, to better understand the incidence of EMD in patients with newly diagnosed B-ALL, data from the E1910 randomized phase 3 trial for adults with B-ALL were incorporated (ClinicalTrials.gov identifier: NCT02003222).
Objectives
The primary objective was to determine the incidence of EMD in patients with R/R CRLF2r B-ALL at diagnosis and relapse. Secondary objectives included describing sites of EMD, evaluating differences between patients with CRLF2r EMD and non-CRLF2r EMD, and describing additional cytogenetic/molecular findings. Exploratory objectives included assessing the incidence of EMD in adults with newly diagnosed CRLF2r B-ALL.
Statistical analysis
Analysis was primarily descriptive. Patient characteristics were presented as medians with ranges for continuous variables and frequency with proportion for categorical variables. Categorical variables were compared using nonparametric tests, including Fisher exact tests, and continuous variables with Mann-Whitney in GraphPad Prism (version 9.3.1). A P value <.05 was considered statistically significant.
Results
Patient characteristics
A total of 179 patients with R/R B-ALL were evaluated. Two patients were excluded: one with Ph-like B-ALL with unconfirmed CRLF2r and another with a lack of available data. Among the 177 patients, 21 (11.9%) had CRLF2r B-ALL, and 156 (88.1%) had non-CRLF2r B-ALL (supplemental Table 1; Figure 1A). Baseline characteristics of patients were largely similar between those with and without CRLF2r. However, patients with CRLF2r were generally older at diagnosis (median age, 18.8 vs 10.0 years; P < .0001), more likely to be male (90.5% vs 66%; P = .023), and of Hispanic ethnicity (61.9% vs 31.4%; P = .0142).
Figure 1.
Overall study design. CONSORT diagram summarizing cohort selection. Of the 177 patients enrolled in early-phase CAR T-cell clinical trials, 21 had CRLF2r and were included in the final analysis. Among 21 patients, 8 (38.1%) had EMD at diagnosis, and 17 (80.9%) had EMD at relapse.
The initial response to induction therapy was similar between the 2 groups, with complete remission with MRD negativity achieved in 56.5% of all the patients (52.4% in CRLF2r vs 57.1% in non-CRLF2r; P = .81). In the R/R setting, both groups were comparably heavily pretreated with a median of 4 prior therapies (range, 1-14). Prior therapy exposure, including hematopoietic cell transplantation (HCT), CAR T-cell therapy, and inotuzumab, was generally similar. However, a higher proportion of patients with CRLF2r received blinatumomab than those without CRLF2r (14 patients [66.7%] vs 53 [34%]; P = .007).
Patient stratification based on evidence of EMD
Of the 177 patients, 97 (54.8%) had EMD at any time point (Table 1), 19 (19.6%) of whom had CRLF2r. Collectively, 19 of the 21 patients (90.5%) with CRLF2r had EMD at some time point; in contrast, 78 of the 156 patients (50%) without CRLF2r ever had EMD (P = .0003; Figure 2A).
Table 1.
Demographic characteristics of patients with EMD at any time point, comparing those with and without CRLF2r
| Baseline characteristics of patients with EMD | All patients with EMD (n = 97) | With CRLF2r (n = 19) | Non-CRLF2r (n = 78) | P value |
|---|---|---|---|---|
| Age at diagnosis, median (range), y | 10.9 (1.2-37) | 18.8 (5.4-33.8) | 10 (1.2-37) | .0001 |
| Sex, n (%) | .0198 | |||
| Male | 70 (72.2) | 18 (94.7) | 52 (66.7) | |
| Female | 27 (27.8) | 1 (5.3) | 26 (33.4) | |
| Ethnicity, n (%) | .0698 | |||
| Non-Hispanic | 58 (59.8) | 8 (42.1) | 50 (64.1) | |
| Hispanic | 37 (38.1) | 11 (57.9) | 26 (33.3) | |
| Not reported | 2 (2.1) | 0 | 2 (2.6) | |
| EMD at initial diagnosis∗, n (%) | 30 (30.9) | 8 (42.1) | 22 (28.2) | .2740 |
| CNS-2 | 16 (16.5) | 3 (15.8) | 13 (16.7) | |
| CNS-3 | 11 (11.3) | 4 (21) | 7 (9) | |
| Non-CNS | 6† (6.2) | 1 (5.3) | 5 (6.4) | |
| CNS and non-CNS EMD at initial diagnosis, n (%) | 2 (2.1) | 0 | 2 (2.5) | |
| Initial response, n (%) | ||||
| CR MRD negative | 60 (61.9) | 11 (57.9) | 49 (62.8) | .7937 |
| CR MRD positive | 24 (24.7) | 5 (26.3) | 19 (24.3) | >.9999 |
| Unknown MRD | 4 (4.1) | 1 (5.3) | 3 (3.9) | |
| Refractory | 9 (9.3) | 2 (10.5) | 7 (9) | >.9999 |
| No. of prior therapies, excluding HCT, median (range) | 4 (2-11) | 4 (2-7) | 4 (2-11) | |
| 2-3 prior lines | 26 (25.3%) | 5 (26.3%) | 21 (26.9%) | |
| >3 prior lines | 71 (74.7%) | 14 (73.7%) | 57 (73.1%) | |
| Prior HCT, n (%) | 53 (54.6) | 7 (36.8) | 46 (59) | .1223 |
| Prior blinatumomab, n (%) | 43 (44.3) | 13 (68.4) | 30 (38.5) | .0223 |
| Prior inotuzumab, n (%) | 19 (19.6) | 6 (31.6) | 13 (16.6) | .1948 |
| Prior CAR, n (%) | 38 (39.2) | 4 (21.1) | 34 (43.6) | .1143 |
EMD status is unknown in 1 patient (no CSF and/or positron emission tomography reports were available).
Six patients had BM and non-CNS EMD at diagnosis (testes, sinus mass, skin, spine, kidney, left para-aortic lesion from lumbar spine, and vitreous). Three from those without CRLF2r had concomitant CNS involvement.
Figure 2.
Characterization and incidence of extramedullary disease in patients with and without CRLF2r. (A) Incidence of EMD in patients with CRLF2r vs non-CRLF2r. Patients with CRLF2r showed a higher rate of EMD than those without CRLF2r (90.5% vs 50%; P = .0003). (B) Incidence of EMD at diagnosis. Eight of the 21 patients (38.1%) with CRLF2r had EMD at diagnosis. In those without CRLF2r (n = 156), 22 patients (14.1%) had EMD at diagnosis (P = .0114). (C) Manifestations of CNS and non-CNS EMD in patients with CRLF2r at relapse.
In the CRLF2r cohort (n = 21), EMD was seen at diagnosis in 8 patients (38.1%), at any relapse time point in 17 patients (80.9%), and at both diagnosis and relapse in 6 patients (28.6%; Figure 1A; supplemental Table 2). Only 2 patients (9.5%) remained without EMD at diagnosis or relapse. No patients with CRLF2r in this cohort had constitutional trisomy 21. Additional molecular aberrations included JAK1, JAK2, NRAS, CDKN2A, CDKN2B, IDH1, STAT3, KRAS, PTPN11, and KMT2D mutations (supplemental Table 3).
In those without CRLF2r (n = 156), 22 patients (14.1%) had EMD at diagnosis, 69 patients (44.2%) at any relapse time point, and 13 patients (8.3%) at both diagnosis and relapse. In total, 78 patients (50%) had no evidence of EMD at diagnosis or relapse. Molecular and/or cytogenetic characteristics varied substantially without a clear association with EMD (supplemental Figure 1).
EMD at diagnosis
At initial presentation, 30 patients (16.9%) of the entire cohort had EMD at diagnosis. Among patients with CRLF2r, 8 (38.1%) had EMD. Of these, 7 patients (87.5%) had CNS EMD (3 CNS-2 and 4 CNS-3), and 1 patient had involvement of a left para-aortic lymph node. Imaging for non-CNS EMD was performed in a single patient at diagnosis because of lumbar and abdominal pain. The remaining patients (n = 7) were evaluated exclusively for CNS involvement; therefore, the presence of subclinical non-CNS EMD was unknown.
In those without CRLF2r (n = 156), 22 patients (14.1%) had EMD at diagnosis compared with 38.1% in those with CRLF2r (P = .0114; Figure 2B). Twenty patients (12.8%) had CNS EMD (13 CNS-2 and 7 CNS-3), and 5 (3.2%) had non-CNS EMD involving sinus mass, testes, skin, thoracic spinal canal, right kidney, and vitreous deposit. Indication for imaging for non-CNS EMD was documented in 3 of the 5 patients and was prompted by physical examination findings, which raised concern for non-CNS EMD. For 1 patient with kidney involvement at diagnosis, the indication for imaging was unknown.
EMD at relapse
At relapse, encompassing any relapse time point, 17 patients (80.9%) with CRLF2r had EMD: 5 (23.8%) had CNS EMD, 8 (38.1%) had non-CNS EMD, and 4 (19%) had both (Table 2). This included 6 patients who had EMD at diagnosis. BM involvement at first relapse was seen in 15 of the 21 patients (71.4%). Clinical sites of non-CNS EMD at relapse varied widely and are summarized in Figure 2C. Detailed individual presentations, along with characterization of molecular and cytogenetic findings, are described in Table 2.
Table 2.
Sites of EMD in those with CRLF2r at diagnosis and after relapse, stratified by patient
| Patient | Disease time point | EMD at disease time point | Concomitant BM disease | Immediate prior therapy |
|---|---|---|---|---|
| 1 | Diagnosis | None | Yes | N/A |
| First relapse | CNS | Yes | HCT | |
| 2 | Diagnosis | None | Yes | N/A |
| First relapse | CNS and VI nerve palsy | Yes∗ | AIEOP-BFM 2009 | |
| Second relapse | Testicular | No | HCT | |
| Third relapse | CNS, VI nerve palsy, testicular, and a mass in the left periaortic lymph nodes | No | 6-mercaptopurine, vincristine, and DLI | |
| 3 | Diagnosis | None | Yes | N/A |
| First relapse | CNS-2 | Yes | Blinatumomab | |
| 4 | Diagnosis | None | Yes | N/A |
| First relapse | CNS and extra-axial mass right frontal sinus and skull chloroma | Yes | HCT | |
| Second relapse | CNS | Yes∗ | 6-mercaptopurine, methotrexate, and prednisone | |
| 5 | Diagnosis | None | Yes | N/A |
| First relapse | CNS | Yes | Cytarabine/Cy | |
| 6 | Diagnosis | None | Yes | N/A |
| First relapse | Para-aortic lymph node and pericaval lymph node | Yes | Anti-CD19 CAR; second infusion | |
| 7 | Diagnosis | CNS | Yes | N/A |
| Relapse | None | N/A | N/A | |
| 8 | Diagnosis | None | Yes | N/A |
| Relapse | None | N/A | N/A | |
| 9 | Diagnosis | None | Yes | N/A |
| First relapse | CNS and parieto-occipital extra-axial mass | No | HCT | |
| 10 | Diagnosis | Left para-aortic lesion from lumbar spine L2 to L5 | Yes | N/A |
| First relapse | Kidney, gastrohepatic lymph node, and bulky inguinal lymph node | Yes∗ | Nivolumab and blinatumomab | |
| Second relapse | Kidney and liver | Yes∗ | Inotuzumab | |
| 11 | Diagnosis | None | Yes | N/A |
| First relapse | Parotid gland | Yes | Vincristine, prednisone, L-asparaginase, daunorubicin, and IT MTX | |
| 12 | Diagnosis | None | Yes | N/A |
| First relapse | CNS | Yes | Fludarabine, mitoxantrone, cytarabine, and filgrastim | |
| 13 | Diagnosis | None | Yes | N/A |
| Relapse | None | Yes | N/A | |
| 14 | Diagnosis | None | Yes | N/A |
| First relapse | CNS, kidney, and bulky retroperitoneal lymphadenopathy encasing the abdominal aorta | Yes | Vincristine, daunorubicin, Peg asparaginase, cytarabine, 6-mercaptopurine, and methotrexate | |
| Second relapse | CNS | No | Inotuzumab | |
| 15 | Diagnosis | None | Yes | None |
| First relapse | Spleen and peripancreatic nodes | Yes | FLAG-IDA | |
| 16 | Diagnosis | None | Yes | N/A |
| First relapse | Optic nerve | Yes∗ | Vincristine, dexamethasone, 6-mercaptopurine, and methotrexate | |
| 17 | Diagnosis | CNS | Yes | N/A |
| First relapse | None | N/A | N/A | |
| 18 | Diagnosis | None | Yes | N/A |
| First relapse | None | N/A | N/A | |
| 19 | Diagnosis | CNS | Yes | N/A |
| First relapse | Liver, kidney, and testicular | Yes | HCT | |
| Second relapse | Kidney, testicular, and retroperitoneal nodes | No | High-dose MTX | |
| 20 | Diagnosis | CNS | Yes | N/A |
| First relapse | Supraclavicular lymph node | No | Blinatumomab | |
| Second relapse | Left upper neck carotid chain, anterior inferior to right parotid gland, right mandibular region, and mediastinal soft tissue involving paratracheal regions and encircling left bronchus | Yes | Cy and cytarabine | |
| 21 | Diagnosis | None | Yes | N/A |
| First relapse | Retroperitoneal lymph nodes, right iliac chain, inguinal lymph node, and subcutaneous mass over the lower hip | Yes | HCT | |
| Second relapse | cervical lymph nodes, confirmed with biopsy | Yes | CD19 CAR |
Cy, cyclophosphamide; DLI, donor lymphocyte infusion; FLAG-IDA, fludarabine, cytarabine, granulocyte colony-stimulating factor, and idarubicin; MTX, methotrexate; N/A, not applicable; IT, intrathecal.
MRD positivity at relapse.
In those without CRLF2r, 69 patients (44.2%) had EMD at any relapse time point: 18 had CNS EMD (11.5%; 7 CNS-2 and 11 CNS-3), 37 (23.7%) had non-CNS EMD, and 14 (9%) had both. Clinical sites of EMD are summarized in supplemental Figure 2.
Subanalyses of patients who received HCT
Ninety patients received allogeneic HCT across a wide range of institutions, donors, and conditioning regimens (Figure 3A). Among the 8 patients with CRLF2r who underwent HCT, 6 (75%) relapsed with EMD, including 3 with a history of CNS EMD at diagnosis. Of the 6 relapses, 2 were isolated EMD: one with CNS involvement and a parieto-occipital extra-axial mass. Other patients had concurrent EMD and marrow relapse, including 1 patient with testicular involvement. For patients without CRLF2r who underwent HCT, 26 of the 82 patients (31.7%) who underwent transplant relapsed with EMD (Figure 3B). Only 1 patient had isolated CNS EMD, with EMD and marrow relapse in all other patients. Overall, the incidence of post-HCT EMD was higher in those with CRLF2r than those without CRLF2r (75% vs 32%; P = .002; Figure 3B).
Figure 3.
Incidence of extramedullary disease after HCT and blinatumomab according to CRLF2r status. (A) Number of patients with and without CRLF2r who underwent HCT in each group. (B) Incidence of EMD after HCT in patients with and without CRLF2r. Ninety patients received HCT. Among the 8 patients with CRLF2r, 6 (75%) relapsed with EMD. In those without CRLF2r, 26 of the 82 patients who underwent transplant (31.7%) relapsed with EMD (P = .002). (C) Number of patients treated with blinatumomab for relapse. Sixty-seven patients received blinatumomab for relapse: 14 with CRLF2r (median, 1 cycle; range, 0.1-3) and 53 without CRLF2r (median, 2 cycles; range, 1-5; P = .007). (D) Incidence of EMD after blinatumomab. In those with CRLF2r treated with blinatumomab, 4 of the 14 patients (28.6%) had EMD after blinatumomab relapse, including 1 case of isolated EMD involving the spleen, confirmed with biopsy tissue. In patients without CRLF2r, 10 of the 53 patients (18.9%) had EMD (P = .46). blina, blinatumomab.
Subanalyses of patients who received blinatumomab or inotuzumab for relapsed disease
Given the increasing use of antigen-targeted therapies in the current era, the incidence of EMD by blinatumomab and inotuzumab exposure was assessed. Among the 177 patients, 67 (37.8%) received blinatumomab for relapse. Blinatumomab exposure differed by CRLF2r status, occurring in 14 of the 21 patients with CRLF2r (66.7%; median, 1 cycle; range, 3 days to 3 cycles) and in 53 of the 156 patients without CRLF2r (34.0%; median, 2 cycles; range, 1-5; P = .007; Figure 3C). The timing of blinatumomab in relation to prior HCT was not available. Two patients in the CRLF2r cohort did not receive a full cycle: one stopped blinatumomab after 16 days because of progressive disease and the other after 3 days because of grade 3 neurotoxicity. Rates of relapse after blinatumomab were not substantially different: among the 14 patients with CRLF2r treated with blinatumomab, 4 (28.6%) had EMD relapse or progression after blinatumomab, one of whom had isolated splenic EMD confirmed with biopsy. In those without CRLF2r, 10 of the 53 patients (18.9%) progressed or relapsed with EMD (P = .46; Figure 3D). One patient had isolated CNS EMD. The EMD sites of relapse after blinatumomab are summarized in supplemental Figure 3.
Inotuzumab was administered for relapse in 36 patients (30.3%), including 7 patients (33%) with CRLF2r and 29 without CRLF2r (18.6%). Among patients with CRLF2r, 2 of the 7 (28.6%) had EMD relapse, including 1 case of isolated CNS relapse. Similarly, 7 of the 29 patients (24.1%) without CRLF2r relapsed with EMD. There was no significant difference in the rate of EMD relapse between groups (P > .99).
EMD in adults with newly diagnosed B-ALL
The phase 3 randomized E1910 study enrolled 481 adults with a median age of 51 years (range, 30-70) with newly diagnosed B-ALL8; 67 (13.9%) had EMD, and 66 (13.7%) had CRLF2r. In the 66 patients with CRLF2r, 19.7% had EMD. In the 415 patients without CRLF2r, 54 (13.0%) had EMD (P = .17; odds ratio, 1.64; 95% confidence interval, 0.81-3.22). By disease sites using routine assessment without incorporation of any systematic screening by imaging, 12.1% of the patients with CRLF2r and 6.5% of the patients with non-CRLF2r alterations had CNS disease; 1.5% of the patients with CRLF2r and 0.2% of the patients with non-CRLF2r alterations had testicular disease; finally, 4.5% of the patients with CRLF2r and 6% of the patients with non-CRLF2r alterations had lymphomatous EMD involvement.
Discussion
In this single-center retrospective analysis of 177 patients with R/R B-ALL, we comprehensively characterized the incidence of EMD in patients with and without CRLF2r. With CRLF2r reported in ∼5% to 10% of patients with pediatric B-ALL at diagnosis and in 10% to 15% of patients lacking other recurrent genetic alterations,1,3,9 the correlation between CRLF2r and EMD is not well characterized. In our cohort, the relative proportion of patients with CRLF2r B-ALL who had EMD at diagnosis (38.1%) was higher than the 5% to 10% incidence previously reported in newly diagnosed B-ALL.10,11 Interestingly, in the E1910 cohort of adults with newly diagnosed BCR-ABL–negative B-ALL,8 the overall incidence of EMD was substantial, with a nonsignificant trend toward higher incidence of EMD in patients with CRLF2r, particularly in the CNS.
At relapse, 80.9% of the patients with CRLF2r had EMD, which is notably higher than the 15% to 25% range of combined medullary/EMD disease reported in previous studies.12 Collectively, the rate of EMD at any time point in our CRLF2r cohort was substantially higher than in patients without CRLF2r. Although CRLF2 expression is commonly associated with JAK2 mutation, our cohort, comprising heavily pretreated patients, exhibited heterogeneous concomitant cytogenetics and molecular abnormalities (Table 2), with an unclear relationship to EMD. This heterogeneity limited mechanistic inferences, and the contribution of these additional comutations remains uncertain.
EMD is a known cause of treatment failure after HCT.13,14 Patients who undergo HCT with unrecognized EMD may have a higher risk of relapse after transplant,15 and a history of EMD might be an independent risk factor for post-HCT EMD.16 A prior study evaluating 268 adult patients undergoing allogeneic HCT demonstrated that ∼14% experienced EMD relapse after HCT. Among these, 6.4% had isolated EMD, involving either CNS or non-CNS sites, whereas 8.2% had combined BM and EMD relapse, potentially representing cases with undetected non-CNS disease before transplantation.16 In our cohort, 6 of the 21 patients (28.6%) with CRLF2r B-ALL relapsed after HCT: 4 (19%) had combined medullary/EMD, and 2 (10%) had both isolated CNS and non-CNS EMD involvement. This raises the concern that the higher rate of EMD in patients with CRLF2r could potentially reflect undiagnosed EMD before HCT.
Similar to the post-HCT setting, prior reports suggest that EMD involvement before blinatumomab salvage therapy has been associated with increased risk of EMD relapse, especially CNS EMD.14,17 Additionally, there remains a concern that neither blinatumomab nor inotuzumab effectively prevents or eradicates EMD.18, 19, 20 In our study, 5 patients (23.8%) relapsed with EMD after blinatumomab and/or inotuzumab (administered for marrow relapse). Three (17.6%) had a combined BM and EMD relapse: 2 with non-CNS and 1 with CNS EMD, whereas 2 patients developed isolated EMD disease after blinatumomab and inotuzumab (Table 2). Although a higher proportion of patients with CRLF2r received blinatumomab, an association with higher rates of EMD relapse was not appreciated relative to those without CRLF2r receiving blinatumomab. Additional studies with larger patient numbers will be essential to understanding the role of novel immunotherapies in EMD risk.
Beyond the inherent limitations of retrospective studies, including missing clinical data information and incomplete baseline variables, such as initial NCI risk classification, additional challenges remain. A major constraint is the inconsistent documentation of EMD at diagnosis, relevant to all patients. Importantly, this reflects variability in clinical practice alongside a lack of routine screening for EMD beyond the CNS or testes. Selection bias for patients with R/R disease, as reflected in our cohort, and more comprehensive screening for trial enrollment may have skewed the results for the higher incidence of EMD overall, but the approach toward detection of EMD was consistent across all patients, in whom baseline characteristics were relatively comparable, with the exception of age and ethnicity. In this regard, although our CRLF2r population was skewed toward being older and more frequently Hispanic, both of these features are associated with CRLF2r.21, 22, 23
To our knowledge, we report on the first characterization of EMD in patients with CRLF2r B-ALL. Although limited by factors inherent to retrospective analyses, our findings nonetheless suggest a potential association between CRLF2r and EMD that warrants further study. Whether this will be applicable to other Ph-like subtypes also warrants exploration. Recognition of this potential relationship, if validated in larger cohorts, has important diagnostic and therapeutic implications because it may influence both the initial evaluation and inform therapeutic decision-making. Enhanced and potentially routine screening for EMD could be incorporated into standard practice, particularly for patients with R/R disease or in the peri-HCT period. More accurate identification of EMD may also refine therapeutic decisions, including consideration of focal radiation for localized lesions or prioritizing therapeutic approaches with established EMD efficacy. Additionally, to raise awareness about the presence of non-CNS EMD in patients with B-ALL, establishing systematic approaches to assess for non-CNS EMD in high-risk patients is needed.
Conflict-of-interest disclosure: N.N.S. reports research funding from Lentigen, Vor Bio, and CARGO Therapeutics; has attended advisory board meetings (no honoraria) for Vor, ImmunoACT, and Sobi; and received royalties from the National Institutes of Health for the development of CD22 chimeric antigen receptor T cells. The remaining authors declare no competing financial interests.
The current affiliation for H.S. is Novartis Pharmaceuticals, East Hanover, NJ.
Acknowledgments
This work was supported in part by the Intramural Research Program, Center for Cancer Research, National Cancer Institute, and National Institutes of Health (NIH) Clinical Center, National Institutes of Health (ZIA BC 011823; N.N.S). Efforts from the E1910 study were supported under award numbers U10CA180820, U10CA180794, UG1CA189859, UG1CA232760, UG1CA233277, P30CA021765, and R35CA197695 from the National Institute of General Medical Sciences of the NIH (P50GM115279) and in part by Amgen.
The contributions of the NIH authors were made as part of their official duties as NIH federal employees, are in compliance with agency policy requirements, and are considered Works of the US Government. However, the findings and conclusions presented in this article are those of the author(s) and do not necessarily reflect the views of the NIH or the US Department of Health and Human Services.
Authorship
Contribution: F.G. and N.N.S. designed and conceived the study concept, conducted primary analysis, and wrote the first draft of the manuscript; B.Y., S.K.S., H.S., and G.L. contributed data and provided patient care and critical input into data analysis and writing; C.O. and B.A. contributed to the data analysis and writing; E.B.T. provided image radiology review; C.M.Y. and H.-W.W. performed and analyzed flow cytometry; Z.S., S.L., R.M., E.P., C.M., and M.L. provided additional input from the E1910 data set, providing clinical data for analysis; and all authors contributed significantly to the final version of the manuscript and agreed to its submission.
Footnotes
Deidentified participant data that support the findings of this article will be shared on an approved written request to researchers who provide a methodologically sound proposal for the purposes of achieving specific aims outlined in that proposal. To gain access, data requesters will need to sign a data access agreement and confirm that data will only be used for the agreed-upon purpose for which access was granted. Requests can be directed to the corresponding author, Nirali N. Shah (nirali.shah@nih.gov).
The full-text version of this article contains a data supplement.
Supplementary Material
Supplemental Figure 1.
Supplemental Figure 2.
Supplemental Figure 3.
References
- 1.Roll JD, Reuther GW. CRLF2 and JAK2 in B-progenitor acute lymphoblastic leukemia: a novel association in oncogenesis. Cancer Res. 2010;70(19):7347–7352. doi: 10.1158/0008-5472.CAN-10-1528. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Silbert SK, Scanlon S, Wang HW, et al. CRLF2-rearranged B-cell ALL with extramedullary lineage switch to AML following CD19-targeted therapy. J Immunother Cancer. 2024;12(10) doi: 10.1136/jitc-2024-009499. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Tran TH, Tasian SK. How I treat Philadelphia chromosome-like acute lymphoblastic leukemia in children, adolescents, and young adults. Blood. 2025;145(1):20–34. doi: 10.1182/blood.2023023153. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Thastrup M, Duguid A, Mirian C, Schmiegelow K, Halsey C. Central nervous system involvement in childhood acute lymphoblastic leukemia: challenges and solutions. Leukemia. 2022;36(12):2751–2768. doi: 10.1038/s41375-022-01714-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Pui CH, Howard SC. Current management and challenges of malignant disease in the CNS in paediatric leukaemia. Lancet Oncol. 2008;9(3):257–268. doi: 10.1016/S1470-2045(08)70070-6. [DOI] [PubMed] [Google Scholar]
- 6.Lenk L, Alsadeq A, Schewe DM. Involvement of the central nervous system in acute lymphoblastic leukemia: opinions on molecular mechanisms and clinical implications based on recent data. Cancer Metastasis Rev. 2020;39(1):173–187. doi: 10.1007/s10555-020-09848-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Shahriari M, Shakibazad N, Haghpanah S, Ghasemi K. Extramedullary manifestations in acute lymphoblastic leukemia in children: a systematic review and guideline-based approach of treatment. Am J Blood Res. 2020;10(6):360–374. [PMC free article] [PubMed] [Google Scholar]
- 8.Litzow MR, Sun Z, Mattison RJ, et al. Blinatumomab for MRD-negative acute lymphoblastic leukemia in adults. N Engl J Med. 2024;391(4):320–333. doi: 10.1056/NEJMoa2312948. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Tasian SK, Loh ML, Hunger SP. Philadelphia chromosome-like acute lymphoblastic leukemia. Blood. 2017;130(19):2064–2072. doi: 10.1182/blood-2017-06-743252. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Holland EM, Yates B, Ling A, et al. Characterization of extramedullary disease in B-ALL and response to CAR T-cell therapy. Blood Adv. 2022;6(7):2167–2182. doi: 10.1182/bloodadvances.2021006035. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Geethakumari PR, Hoffmann MS, Pemmaraju N, et al. Extramedullary B lymphoblastic leukemia/lymphoma (B-ALL/B-LBL): a diagnostic challenge. Clin Lymphoma Myeloma Leuk. 2014;14(4):e115–e118. doi: 10.1016/j.clml.2014.01.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Cooper SL, Brown PA. Treatment of pediatric acute lymphoblastic leukemia. Pediatr Clin North Am. 2015;62(1):61–73. doi: 10.1016/j.pcl.2014.09.006. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Agrawal V, Aldoss I, Pullarkat V. Extramedullary sites of disease in B-cell acute lymphoblastic leukemia: incidence, biology, and treatment. Leuk Lymphoma. 2025;66(12):2168–2179. doi: 10.1080/10428194.2025.2541024. [DOI] [PubMed] [Google Scholar]
- 14.Riviello-Goya S, Acosta-Medina AA, Inclan-Alarcon SI, Garcia-Miranda S, Bourlon C. Isolated extramedullary relapse in acute lymphoblastic leukemia: what can we do before and after transplant? Oncology (Williston Park) 2020;34(2):39–43. [PubMed] [Google Scholar]
- 15.Shem-Tov N, Saraceni F, Danylesko I, et al. Isolated extramedullary relapse of acute leukemia after allogeneic stem cell transplantation: different kinetics and better prognosis than systemic relapse. Biol Blood Marrow Transplant. 2017;23(7):1087–1094. doi: 10.1016/j.bbmt.2017.03.023. [DOI] [PubMed] [Google Scholar]
- 16.Yu J, Ge X, Luo Y, et al. Incidence, risk factors and outcome of extramedullary relapse after allogeneic hematopoietic stem cell transplantation in patients with adult acute lymphoblastic leukemia. Ann Hematol. 2020;99(11):2639–2648. doi: 10.1007/s00277-020-04199-9. [DOI] [PubMed] [Google Scholar]
- 17.Tan JY, Halim NAA, Ang CH, et al. Outcomes of B-cell acute lymphoblastic leukemia patients who relapse after blinatumomab salvage therapy - a Tertiary Centre experience. Blood. 2023;142(suppl 1):5816. [Google Scholar]
- 18.Aldoss I, Otoukesh S, Zhang J, et al. Extramedullary disease relapse and progression after blinatumomab therapy for treatment of acute lymphoblastic leukemia. Cancer. 2022;128(3):529–535. doi: 10.1002/cncr.33967. [DOI] [PubMed] [Google Scholar]
- 19.Johns C, Erickson C, Jacobs A, et al. Case report: sub-clinical extramedullary B-ALL in the setting of relapse following targeted therapy. Front Immunol. 2024;15 doi: 10.3389/fimmu.2024.1423487. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Kayser S, Sartor C, Luskin MR, et al. Outcome of relapsed or refractory acute B-lymphoblastic leukemia patients and BCR-ABL-positive blast cell crisis of B-lymphoid lineage with extramedullary disease receiving inotuzumab ozogamicin. Haematologica. 2022;107(9):2064–2071. doi: 10.3324/haematol.2021.280433. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Juarez-Velazquez MDR, Moreno-Lorenzana DL, Martinez Anaya DA, et al. High occurrence of CRLF2 abnormalities in Mexican children with B-cell acute lymphoblastic leukemia. Cytokine. 2022;155 doi: 10.1016/j.cyto.2022.155896. [DOI] [PubMed] [Google Scholar]
- 22.Raca G, Abdel-Azim H, Yue F, et al. Increased Incidence of IKZF1 deletions and IGH-CRLF2 translocations in B-ALL of Hispanic/Latino children-a novel health disparity. Leukemia. 2021;35(8):2399–2402. doi: 10.1038/s41375-021-01133-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Tasian SK, Hurtz C, Wertheim GB, et al. High incidence of Philadelphia chromosome-like acute lymphoblastic leukemia in older adults with B-ALL. Leukemia. 2017;31(4):981–984. doi: 10.1038/leu.2016.375. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.