CNS involvement by North American-ATLL (NA-ATLL) is associated with discrete patterns and molecular profile involving XPO1 E571 and KLF2/PI3KCD in selected cases

8063

Background: Information on central nervous system (CNS) involvement with NA-ATLL is limited. In this study, we describe CNS involvement in ATLL patients at a tertiary hospital in New York City. Methods: We considered CNS involvement if one of the following criteria was met 1) cerebrospinal fluid (CSF) cytology or flow cytometry was positive 2) CNS imaging was positive for disease involvement or 3) Physical exam findings were compatible with neurologic involvement. Results: Of 94 NA-ATLL patients, 21 (22.3%) had CNS involvement by ATLL. CSF was involved in 13/21 and 5/21 patients at diagnosis and relapse respectively. At diagnosis, MRI showed CNS involvement in brain and spine in 5/21 (24%) and 3/ 21 (14%) cases respectively. At relapse, imaging revealed brain and spine involvement in 2 patients each. Neurological exam was abnormal in 7 (33%) and 3 (14%) cases at diagnosis and relapse respectively. Next generation exon targeted sequencing was performed in 9 cases. Table shows the mutations (mtn) and functional groups with frequencies. XPO1 E571K mutation was present in 2 patients with extensive CNS disease and refractory to conventional treatment with an overall survival (OS) of 2 months. To our knowledge, this is the first time that XPO1 E571K is reported in a T-Cell malignancy. We also describe here a second set of mutations with CNS involvement (KLF2 and PI3KCD) in 2 patients. Median OS was 8.5 months, Median RFS was 6.5 months in our series. In most cases, the lymphomatous phenotype appeared to have direct mass-like extension (5/21) with several cases of accompanying osteolytic spine or skull lesions, whereas cases with hematogenous involvement tends to spread to the CSF by traversing the brain blood barrier. Conclusions: In this report we describe patterns of CNS involvement in ATLL and the associated mtns. We also describe two unique cases of XPO1E571K mtn in a TCL. [Table: see text]

Targeted combination therapy with fulvestrant (FUL) for second-line (2L) treatment of hormone receptor-positive (HR+) advanced breast cancer (ABC)

e12527

Background: FUL is the recommended 2L treatment for patients whose HR+ ABC progressed after aromatase inhibitor (AI) therapy. In first line ABC adding targeted therapy, eg. cyclin-dependent kinase 4/6 inhibitors (CDK4/6i) palbociclib or ribociclib, or mammalian target of rapamycin inhibitor (mTORi) everolimus (EVE), to endocrine therapy (ET) has shown superior efficacy vs ET alone. The use of similar strategies to delay disease progression on ET in the 2L setting is an area of active research. Methods: PubMed and ClinicalTrials.gov were searched for trials investigating FUL + targeted therapies in 2L HR+ ABC. Search terms: (advanced OR metastatic) AND (breast cancer) AND (FUL OR faslodex) AND (2L OR relapse OR refractory OR resistant OR progression). Efficacy, adverse events (AEs) and quality of life were assessed. Results: 28 studies of FUL + targeted therapies in 2L ABC were found. Key randomized trials include 8 studies exploring FUL + CDK4/6i: palbociclib, ribociclib or abemaciclib. Palbociclib + FUL significantly improved progression-free survival (PFS) vs FUL in 2L HR+ ABC (p < 0.0001; PALOMA-3), AEs were manageable. Assessment of FUL + ribociclib (MONALEESA-3) or FUL + abemaciclib (MONARCH-2) in 2L HR+ ABC is ongoing. Ten studies are evaluating FUL + phosphatidylinositol 3-kinase (PI3K)/AKT/mTORi in 2L HR+ ABC. FUL + EVE significantly prolonged PFS vs FUL (p = 0.02; PrECOG 0102). Two trials evaluated buparlisib (pan-PI3K inhibitor [PI3Ki]) + FUL in HR+ ABC post-AI (BELLE-2) and post-mTORi (BELLE-3). In both trials, buparlisib + FUL improved PFS vs FUL in patients with PIK3CA-mutated tumors, FUL alone led to a poor response in this subgroup. However, pictilisib (pan-PI3Ki) + FUL did not improve PFS vs FUL even in the PIK3CA-mutated subgroup (FERGI). Ongoing phase 3 trials are exploring FUL + alpelisib (α-specific PI3Ki; SOLAR-1) or FUL + taselisib (β-sparing PI3Ki; SANDPIPER) in PIK3CA-mutant HR+ ABC. Pts who have progressed on AI, CDK4/6i or (neo)adjuvant chemotherapy are eligible for these studies. Data on FUL + other targeted therapies will also be discussed. Conclusions: Addition of targeted therapy to FUL demonstrates promising efficacy beyond first line.

Chronic hepatitis C as an adverse prognostic factor in myelodysplastic syndromes

e18557

Background: Chronic hepatitis C virus (HCV) infection has been associated with hematologic malignancies such as B-cell non-Hodgkin lymphoma. In contrast, there is very little known about the relationship between HCV and myeloid malignancies such as myelodysplastic syndromes (MDS) and acute myeloid leukemia (AML). To determine the effect of HCV on MDS/AML clinical outcomes we conducted a retrospective analysis in a large inner city cohort of patients (pts) Methods: 478 pts with MDS and MDS-transformed AML were identified between 1997 and 2016; 13 pts were HCV-positive (HCV+/MDS) and 465 were HCV-negative (HCV-/MDS). Demographics, hematologic parameters, cytogenetics, IPSS-R scores and molecular profiles were compared for both groups (grp). Survival (surv) analysis was done using the Kaplan-Meier method Results: HCV+/MDS had significantly worse surv than HCV-/MDS pts on univariate analysis (UA) (median surv 16 vs 52 months, HR 2.8, 95% CI 1.4-5.5, p < 0.01). Difference remained strongly significant after exclusion of HIV-coinfected pts (HR 2.7, p = 0.02) and pts w/ AML on presentation (HR 3.1, p < 0.01). Cytopenias were worse in HCV+/MDS (mean Hgb level 8.2 vs 9.4 g/dl, p = 0.04; mean plt count 64 vs 139 103/µL, p = 0.03). Average IPSS-R score was higher in HCV+/MDS pts (5.5 vs 3.8, p = 0.02). IDH1 mutation was more prevalent in HCV+/MDS (25% vs 2%, p < 0.01). Differences in age between HCV+/MDS and HCV-/MDS were non-significant (mean 64 vs 69 respectively, p = 0.2). Male-to-female ratio was higher in HCV+/MDS but difference was non-significant (2.3 vs 0.9, p = 0.2). Interestingly, high HCV viral load and cirrhosis did not correlate with poor surv in the HCV+/MDS grp, suggesting deleterious effect occurs even in early stages of HCV infection. Conclusions: Our analysis suggests that chronic HCV infection strongly correlates with worse surv in UA and is associated with lower blood counts and higher IPSS-R scores at the time of diagnosis of MDS. We observed a high rate of MDS refractory to standard therapies and speculate that HCV may affect biology of the disease. Larger studies are warranted to determine the effect of HCV on surv in this population and to provide a rationale for trials of anti-HCV drugs concomitantly with MDS Tx.