Faculty Bibliography

High-dose melphalan (MEL200) followed by autologous stem cell transplantation (ASCT) remains a standard of care for multiple myeloma (MM). Bendamustine induces responses in MM resistant to other alkylators. Our prior Phase I trial adding bendamustine to MEL200 transplant conditioning resulted in no additional toxicity. We now report a single-arm, phase II study that evaluated the efficacy of bendamustine 225 mg/m² with MEL200 conditioning for ASCT in 18 patients with newly diagnosed MM (NDMM) and 17 with relapsed or refractory MM (RRMM). The primary end point was the complete response (CR/sCR) rate at day+ 100. Sample size was determined according to Simon's two-stage design. At stage 1, sixteen patients entered the study. As there were eight patients with CR/sCR, enrollment increased to 28 patients. Sixteen out of the first 28 evaluable patients achieved CR/sCR, meeting the design criteria. Enrollment was then expanded to a total of 35 patients. 51% achieved a CR/sCR. After a median follow-up of 65 months, 21 patients progressed, including 7 deaths. The median PFS for NDMM and RRMM was 48 and 45 months, respectively. Bendamustine/MEL200 conditioning resulted in excellent overall and depth of response as well as PFS, particularly in the RRMM patients, and is worthy of further investigation (NCT00916058).

While the majority of thyroid cancer patients are easily treatable, those with anaplastic or poorly differentiated recurrent thyroid carcinomas have a very poor prognosis with a median survival of less than a year. Previously, we have shown a significant correlation between ICAM-1 overexpression and malignancy in thyroid cancer, and have pioneered the use of ICAM-1 targeted CAR T cells as a novel treatment modality. For clinical translation of this novel modality, we designed CAR T cells possessing micromolar rather than nanomolar affinity to ICAM-1 to avoid cytotoxicity in normal cells with basal levels of ICAM-1 expression. Herein, we report the automated process of CAR T cell manufacturing with CliniMACS Prodigy (Miltenyi Biotec) using cryopreserved peripheral blood leukocytes from apheresis collections. Using Prodigy, thawed leukopak cells were enriched for CD4⁺ and CD8⁺ T cells, subjected to double transduction using lentiviral vector, and expanded in culture for a total of 10 days with a final yield of 2-4 × 10⁹ cells. The resulting CAR T cells were formulated for cryopreservation to be used directly for infusion into patients after thawing with no further processing. We examined cross-reactivity of CAR T cells toward both human and murine ICAM-1 and ICAM-1 expression in human and mouse tissues to demonstrate that both efficacy and on-target, off-tumor toxicity can be studied in our preclinical model. Selective anti-tumor activity in the absence of toxicity provides proof-of-concept that micromolar affinity tuned CAR T cells can be used to target tumors expressing high levels of antigen while avoiding normal tissues expressing basal levels of the same antigen. These studies support the initiation of a phase I study to evaluate the safety and potential efficacy of micromolar affinity tuned CAR T cells against newly diagnosed anaplastic and refractory or recurrent thyroid cancers.

Epstein-Barr virus (EBV) reactivation and post-transplant lymphoproliferative disorders (PTLD) are common and potentially fatal complications after allogeneic transplantation with mismatched donors and T-cell depletion. Haplo-cord transplantation combines a mismatched UCB graft with third-party cells. Conditioning involves thymoglobulin. EBV reactivation and PTLD were common in initial patients. As of March 2017, we administered a prophylactic dose of rituximab 375 mg/m² pre-transplant. Among 147 patients who did not receive rituximab, the cumulative incidence of post-transplant EBV reactivation and of EBV PTLD was 13% and 8%, respectively. Among 51 who received pre-transplant rituximab, the incidences were 2% (p = .0017) and 0% (p = .04), respectively. There was no difference in time to hematopoietic recovery, in the incidence of CMV reactivation, of invasive blood stream infections or of proven or probable invasive fungal infections. Pre-transplant administration of rituximab is an effective and nontoxic intervention that drastically reduces EBV reactivation and PTLD in high-risk patients.

Adoptive immunotherapy has shown efficacy in patients with relapsed/refractory acute myelogenous leukemia (AML). We conducted a prospective evaluation of cord blood (CB)-based adoptive cell therapy following salvage chemotherapy in patients with AML or myelodysplastic syndrome (MDS) and describe the safety and early outcomes of this approach. To enhance the antileukemic effect, we selected CB units (CBUs) with a shared inherited paternal antigen (IPA) and/or noninherited maternal antigen (NIMA) match with the recipients. Furthermore, the CBUs had total nucleated cell (TNC) dose <2.5 × 10⁷/kg and were at least 4/6 HLA-matched with the patients; a higher allele-level match was preferred. Heavily pretreated adult patients with AML/MDS were enrolled. CBU searches were performed for 50 patients. CBUs with shared IPA targets were identified for all, and CBUs with NIMA matches were found for 80%. Twenty-one patients underwent treatment (AML, primary induction failure, n = 8; refractory relapse, n = 10, including 7 recipients of previous allogeneic HSCT; blast crisis chronic myelogenous leukemia, n = 1; MDS, n = 2). Most received combination chemotherapy; those not fit for intensive treatment received a hypomethylating agent. Response was defined as <10% residual blasts in hypocellular bone marrow at approximately 2 weeks after treatment. Ten of the 19 evaluable patients responded, including 5 of the 7 recipients of previous transplant. Response was seen in 4 of 4 patients with full CBU-derived chimerism, 2 of 2 of those with partial, low-level chimerism and 4 of 12 of the recipients with no detectable CBU chimerism. The most common adverse events were infections (bacterial, n = 5; viral, n = 2; fungal, n = 5). Grade IV acute graft-versus-host disease (GVHD) developed in 2 patients with full CBU chimerism; 2 other patients had grade 1 skin GVHD. A total of 11 patients died, 7 from disease recurrence and 4 from infections (1 early death; the other 3 in remission at the time of death). Overall, 12 patients proceeded to allogeneic HSCT; of those, 7 had responded to treatment, 3 had not (and had received additional therapy), and 2 had persistent minimal residual disease. In conclusion, the use of CB as adoptive immunotherapy in combination with salvage chemotherapy for patients with refractory AML/MDS is feasible, can induce disease control, can serve as a bridge to allogeneic HSCT, and has an acceptable incidence of adverse events. Alloreactivity was enhanced through the selection of CBUs targeting a shared IPA and/or NIMA match with the patients. CBUs with lower cell doses, already available in the CB bank and unlikely to be adequate grafts for adult transplants, can be used for cell therapy within a short time frame.

Donor-specific HLA antibodies (DSAs) have been associated with an increased risk of graft failure. To decrease DSA levels and reduce the risk of graft failure in haploidentical cord blood transplantation recipients, we studied the effect of bortezomib (BTZ) and i.v. immune globulin (IVIG) pretransplantation. Between 2012 and 2016, 14 patients with a DSA level >2000 mean fluorescence intensity (MFI) to 1 or more mismatched HLA alleles of haploidentical donors, cord blood donors, or both were treated with BTZ and IVIG. Fourteen patients received a median of 4 doses (range, 2 to 8 doses) of BTZ 1.3 mg/m² and a median total IVIG of 2 g/kg before transplantation. Only 2 of 14 patients attained a reduction in MFI to <2000 with this combination. After additional IVIG (n = 8), rituximab (n = 4), and/or plasmapheresis (n = 11), 12 of 14 patients were desensitized to a DSA level <2000 MFI at the time of engraftment. All obtained initial hematopoietic reconstitution, and no DSA rebound phenomenon was observed. Responders with DSA MFI <2000 to the haploidentical donor by transplantation engrafted at a rate comparable to that of historical controls, whereas engraftment in nonresponders took 3 times as long. BTZ and IVIG alone do not appear sufficient to rapidly induce DSA desensitization, and persistent DSAs to a haploidentical donor lead to delayed count recovery. Our data suggest that additional pretreatment with BTZ and IVIG in combination with the conditioning regimen may help abrogate the rebound phenomenon observed with plasmapheresis.

Limited studies have reported on outcomes for lymphoid malignancy patients receiving alternative donor allogeneic stem cell transplants. We have previously described combining CD34-selected haploidentical grafts with umbilical cord blood (haplo-cord) to accelerate neutrophil and platelet engraftment. Here, we examine the outcome of patients with lymphoid malignancies undergoing haplo-cord transplantation at the University of Chicago and Weill Cornell Medical College. We analyzed 42 lymphoma and chronic lymphoblastic leukemia (CLL) patients who underwent haplo-cord allogeneic stem cell transplantation. Patients underwent transplant for Hodgkin lymphoma (n = 9, 21%), CLL (n = 5, 12%) and non-Hodgkin lymphomas (n = 28, 67%), including 13 T cell lymphomas. Twenty-four patients (52%) had 3 or more lines of therapies. Six (14%) and 1 (2%) patients had prior autologous and allogeneic stem cell transplant, respectively. At the time of transplant 12 patients (29%) were in complete remission, 18 had chemotherapy-sensitive disease, and 12 patients had chemotherapy-resistant disease. Seven (17%), 11 (26%), and 24 (57%) patients had low, intermediate, and high disease risk index before transplant. Comorbidity index was evenly distributed among 3 groups, with 13 (31%), 14 (33%), and 15 (36%) patients scoring 0, 1 to 2, and ≥3. Median age for the cohort was 49 years (range, 23 to 71). All patients received fludarabine/melphalan/antithymocyte globulin conditioning regimen and post-transplant graft-versus-host disease (GVHD) prophylaxis with tacrolimus and mycophenolate mofetil. The median time to neutrophil engraftment was 11 days (range, 9 to 60) and to platelet engraftment 19.5 days (range, 11 to 88). Cumulative incidence of nonrelapse mortality was 11.6% at 100 days and 19 % at one year. Cumulative incidence of relapse was 9.3% at 100 days and 19% at one year. With a median follow-up of survivors of 42 months, the 3-year rates of GVHD relapse free survival, progression-free survival, and overall survival were 53%, 62%, and 65%, respectively, for these patients. Only 8% of the survivors had chronic GVHD. In conclusion, haplo-cord transplantation offers a transplant alternative for patients with recurrent or refractory lymphoid malignancies who lack matching donors. Both neutrophil and platelet count recovery is rapid, nonrelapse mortality is limited, excellent disease control can be achieved, and the incidence of chronic GVHD is limited. Thus, haplo-cord achieves high rates of engraftment and encouraging results.

Administration of granulocyte colony-stimulating factor (G-CSF) after autologous peripheral blood stem cell transplantation (PBSCT) is generally recommended to reduce the duration of severe neutropenia; however, data regarding the optimal timing of G-CSFs post-transplantation are limited and conflicting. This retrospective study was performed at NewYork-Presbyterian/Weill Cornell Medical Center between November 5, 2013, and August 9, 2016, of adult inpatient autologous PBSCT recipients who received G-CSF empirically starting on day +5 (early) versus on those who received G-CSF on day +12 only if absolute neutrophil count (ANC) was <0.5 × 10⁹/L (ANC-driven). G-CSF was dosed at 300 µg in patients weighing <75 kg and 480 µg in those weighing ≥75 kg. One hundred consecutive patients underwent autologous PBSCT using either the early (n = 50) or ANC-driven (n = 50) G-CSF regimen. Patient and transplantation characteristics were comparable in the 2 groups. In the ANC-driven group, 24% (n = 12) received G-CSF on day +12 and 60% (n = 30) started G-CSF earlier due to febrile neutropenia or at the physician's discretion, 6% (n = 3) started after day +12 at the physician's discretion, and 10% (n = 5) did not receive any G-CSF. The median start day of G-CSF therapy was day +10 in the ANC-driven group versus day +5 in the early group (P < .0001). For the primary outcome, the median time to neutrophil engraftment was 12 days (interquartile range [IQR] 11-13 days) in the early group versus 13 days (IQR, 12-14 days) in the ANC-driven group (P = .07). There were no significant between-group differences in time to platelet engraftment, 1-year relapse rate, or 1-year overall survival. The incidence of febrile neutropenia was 74% in the early group versus 90% in the ANC-driven group (P = .04); however, there was no significant between-group difference in the incidence of positive bacterial cultures or transfer to the intensive care unit. The duration of G-CSF administration until neutrophil engraftment was 6 days in the early group versus 3 days in the ANC-driven group (P < .0001). The median duration of post-transplantation hospitalization was 15 days (IQR, 14-19 days) in the early group versus 16 days (IQR, 15-22 days) in the ANC-driven group (P = .28). Our data show that early initiation of G-CSF (on day +5) and ANC-driven initiation of G-CSF following autologous PBSCT were associated with a similar time to neutrophil engraftment, length of stay post-transplantation, and 1-year overall survival.

Inversion of chromosome 16 is a consistent finding in patients with acute myeloid leukemia subtype M4 with eosinophilia, which generates a CBFB-MYH11 fusion gene. Previous studies showed that the interaction between CBFβ-smooth muscle myosin heavy chain (SMMHC; encoded by CBFB-MYH11) and RUNX1 plays a critical role in the pathogenesis of this leukemia. Recently, it was shown that chromodomain helicase DNA-binding protein-7 (CHD7) interacts with RUNX1 and suppresses RUNX1-induced expansion of hematopoietic stem and progenitor cells. These results suggest that CHD7 is also critical for CBFB-MYH11-induced leukemogenesis. To test this hypothesis, we generated Chd7

PURPOSE/OBJECTIVE:DNA methyltransferase 3A (DNMT3A) is one of the commonly mutated genes in acute myelogenous leukemia (AML). Reports on the prognostic significance of DNMT3A mutations have been inconsistent, and most of the data are available only for patients 60 years of age or younger. We hypothesized that this inconsistency is due to an interaction between the dose of anthracycline used in induction therapy and DNMT3A status. We studied whether patients with DNMT3A-mutated AML treated with standard dose anthracyclines had an inferior survival compared with patients with other mutation profiles or those who received high-dose therapy. EXPERIMENTAL DESIGN/METHODS:A total of 152 patients in this retrospective cohort study (median age, 54 years) with de novo AML underwent induction therapy and next-generation sequencing of 33 commonly mutated genes in hematologic malignancies, including DNMT3A, FLT3-ITD, NPM1, and IDH1/2. Cox regression was used to know whether those with DNMT3A mutations who were treated with standard dose anthracycline had inferior survival. RESULTS:DNMT3A mutations, found in 32% of patients, were not associated with an inferior survival. Dose escalation of anthracycline in the induction regimen was associated with improved survival in those with DNMT3A mutations but not those with wild-type DNMT3A. Patients with DNMT3A mutations who received standard dose induction had shorter survival time than other patient groups (10.1 months vs. 19.8 months, P = 0.0129). This relationship remained significant (HR, 1.90; P = 0.006) controlling for multiple variables. CONCLUSIONS:Patients with DNMT3A-mutated AML have an inferior survival when treated with standard-dose anthracycline induction therapy. This group should be considered for high-dose induction therapy.

Mesenchymal stem cells (MSCs) are a heterogeneous population of non-hematopoietic precursor cells predominantly found in the bone marrow. They have been recently reported to home towards the hypoxic tumor microenvironment in vivo. Interleukin-6 is a multifunctional cytokine normally involved in the regulation of the immune and inflammatory response. In addition to its normal function, IL-6 signaling has been implicated in tumorigenesis. Solid tumors develop hypoxia as a result of inadequate O(2) supply. Interestingly, tumor types with increased levels of hypoxia are known to have increased resistance to chemotherapy as well as increased metastatic potential. Here, we present evidence that under hypoxic conditions (1.5% O(2)) breast cancer cells secrete high levels of IL-6, which serve to activate and attract MSCs. We now report that secreted IL-6 acts in a paracrine fashion on MSCs stimulating the activation of both Stat3 and MAPK signaling pathways to enhance migratory potential and cell survival. Inhibition of IL-6 signaling utilizing neutralizing antibodies leads to attenuation of MSC migration. Specifically, increased migration is dependent on IL-6 signaling through the IL-6 receptor. Collectively, our data demonstrate that hypoxic tumor cells specifically recruit MSCs, which through activation of signaling and survival pathways facilitate tumor progression.