Abstract: Introduction: Adult T-cell leukemia/lymphoma (ATL) is an aggressive peripheral T-cell lymphoma caused by human T-cell lymphotropic virus type I. In younger patients, allogeneic hematopoietic stem cell transplantation (HSCT) has been shown to improve prognosis through a graft-versus-ATL (GvATL) effect. On the other hand, elderly patients with ATL are often not amenable to HSCT, and the prognosis is extremely poor with existing chemotherapy. The most widely used chemotherapy for ATL, CHOP (combination of cyclophosphamide-doxorubicin-vincristine-prednisone) and similar chemotherapy, have shown a complete response (CR) rate of about 20% and a long-term prognosis of less than 10%. In a phase II study of mogamulizumab, a humanized defucosylated anti-C-C chemokine receptor 4 monoclonal antibody, in concurrent combination with chemotherapy (mLSG15), the primary endpoint of CR was 52% in the mLSG15 + mogamulizumab group compared to 33% in the mLSG15 alone group. However, no improvement was observed in overall survival (OS) or progression-free survival (PFS). In addition, mLSG15 is a potent chemotherapy and not feasible in majority of elderly patients. Considering the possibility that mogamulizumab may unlock the tumor immune evasion mechanism by removing regulatory T cells and exert an immunological effect like GvATL, we investigated its usefulness as an immunological consolidation therapy following chemotherapy, with the aim of suppressing recurrence. Methods: New-onset treatment-naive ATL patients who were not eligible to HSCT received 3 courses of CHOP (combination of cyclophosphamide div 750 mg/m 2/day1, doxorubicin iv 50 mg/m 2/day1, vincristine iv 1.4 mg/m 2/day1 [max. 2.0 mg/body, and prednisone po/iv 100 mg/body/day1-5) at 21-day intervals, and then the first course of mogamulizumab (1 mg/kg for 8 doses at 2-week intervals) was initiated between the start of the third course of CHOP-21 and day 42. The primary endpoint was OS at months 12. Secondary endpoints were proportion of subjects with PFS, OS, overall response rate (ORR), organ-specific response rate, CR rate and adverse events (AEs) at month 12. Results: A total of 24 patients were enrolled and the median age was 75.5 years (range, 64-85). The median follow-up period was 11.0 (3.0-30.6) months. Five subjects discontinued study treatment before mogamulizumab administration attributable to insufficient effect (n=2), continuation deemed inappropriate (n=2), and start of the next course extended (n=1). ORR was 87.5%. Organ-specific response rates for peripheral blood, other than peripheral blood, target lesion, and skin lesion were 83.3%, 87.5%, 79.2%, and 66.7%, respectively. OS and PFS rates in the full analysis set (FAS) at months 12 were 52.6 (30.9-70.4)% and 26.6 (10.9-45.3)%, respectively. The median survival time (MST) and 80% confidence interval (CI) for OS was 12.1 (5.0 -21 .0) months (Figure 1). Since the lower limit of the 80% CI was below the threshold of 6 months, this treatment was not considered to be effective. However, the point estimate of MST was 12.1 months, which was equivalent to the expected value. One possible reason for the failure to reach statistical significance could be attributable to occurrence of events in the early stage of the protocol regimen. In other words, patients who did not respond to the preceding chemotherapy did not benefit from mogamulizumab, and only patients who managed to overcome chemotherapy benefited from the immune effects of mogamulizumab, resulting in long-term survival (Figure 1). In the long-term follow-up, the MST (95% CI) of the FAS was 12.4 (4.9-21.0) months, and the median OS (95% CI) at 24 months was 25.1 (9.6-44.1) %. AEs were observed in 16 subjects, of which serious AEs (grade 4/5) were observed in 5 subjects (acute respiratory distress syndrome, ventricular fibrillation, heart failure, hyponatremia, and pneumonia). Largely, AEs were resolved/recovered in 14 subjects including 3 subjects with serious AE; 2 subjects resulted in death, which were not considered treatment-related. Conclusions: Mogamulizumab as an immunological consolidation therapy after CHOP 21 might be one of the treatment options for transplant-ineligible elderly patients with previously-untreated ATL, despite not clearly showing improved OS. Further studies are warranted to examine the dosing timing of mogamulizumab to obtain its maximum benefit for prolonged survival. Figure 1 Figure 1. Disclosures Kato: Abbvie: Consultancy, Research Funding; AstraZeneca: Consultancy; Bristol-Myers Squibb: Honoraria, Research Funding; Celgene: Consultancy, Honoraria, Research Funding; Chugai: Consultancy, Honoraria, Research Funding; Daiichi Sankyo: Consultancy, Research Funding; Dainippon-Sumitomo: Honoraria; Eisai: Consultancy, Research Funding; Janssen: Consultancy, Honoraria, Research Funding; Kyowa Kirin: Honoraria, Research Funding; MSD: Honoraria; Mundi: Honoraria; Novartis: Consultancy, Research Funding; Ono: Honoraria, Research Funding. Suzuki: Kyowa-kirin: Honoraria, Research Funding, Speakers Bureau; Chugai: Honoraria, Research Funding, Speakers Bureau; Taiho: Research Funding; Ohtsuka: Honoraria, Research Funding; Takeda: Research Funding, Speakers Bureau; Shionogi: Research Funding; Eisai: Honoraria, Research Funding; Bristol-Meyer Squib: Honoraria; MSD: Honoraria; Janssen: Honoraria; Abbvie: Speakers Bureau; Meiji Seika: Honoraria, Speakers Bureau; Ohtsuka: Honoraria. Akashi: Sumitomo Dainippon Pharma: Consultancy; Kyowa Kirin: Consultancy, Research Funding; Celgene: Research Funding; Astellas Pharma: Research Funding; Shionogi: Research Funding; Asahi Kasei Pharma: Research Funding; Chugai: Research Funding; Bristol-Myers Squibb: Research Funding.

Abstract: Background: Human herpes virus-6 (HHV6)-associated limbic encephalitis/myelitis is rare but life-threatening nervous system complication following allogeneic hematopoietic stem cell transplantation (allo-HSCT). When the allo-HSCT recipients present encephalitis-associated manifestations such as short-term memory dysfunction, disorientation, consciousness disturbance and seizures, it is essential to immediately perform polymerase chain reaction (PCR) analysis for detection of HHV6 viral DNA in the cerebrospinal fluid (CSF) for the definite diagnosis. However, if reactivation of HHV6 is confined to the spinal cord but not cerebrum, these patients lack encephalitis-like symptoms but manifest only sensory nerves-related symptoms such as severe unexplained pain, dysesthesia and prutitus, leading to the delayed examination of HHV6 DNA detection in CSF. As a result, the patients with HHV6-associated myelitis can be sometimes misdiagnosed as having calcineurin inhibitor (CI)-induced pain syndrome (CIPS), since such manifestations are commonly observed in these two syndromes. In this study, we evaluate incidence and clinical features of HHV6 encephalitis/myelitis and CIPS after allo-HSCT. Method: We retrospectively reviewed the medical records of 435 patients who underwent allo-HSCT between 2002 and 2014 in our facility. HHV6-associated encephalitis/myelitis was directly proved when HHV6 DNA was detected by PCR in CSF. For those patients who were unable to undergo lumbar puncture because of severe thrombocytopenia or a deteriorated general condition, we diagnosed HHV6-associated encephalitis/myelitis if they satisfied more than 2 of the following 3 criteria: (1) typical clinical manifestations as described above; (2) detection of HHV6 DNA in peripheral blood; or (3) limbic encephalopathy based on the selective involvement of the medial temporal lobe on magnetic resonance imaging (MRI). CIPS was diagnosed using three factors: (a) typical clinical manifestations including unexplained severe pain and cutaneous pruritus without skin rash; and (b) not detection of HHV6 DNA in CSF; or (c) abnormal findings at X-ray, MRI, or bone scintigraphy at joint of knee and foot. Results: Twenty-five patients were diagnosed as having HHV-6 encephalitis/myelitis with a cumulative incidence of 5.7%. Median onset was on day 19 after transplantation. HHV-6 encephalitis/myelitis was documented in 15 of 99 cord blood transplant (CBT) recipients (15.2%) and 10 of 336 recipients (3.0%) transplanted with bone marrow or peripheral blood stem cells. This result suggests that a higher incidence of HHV-6 encephalitis/myelitis occurring in CBT recipients. Four patients manifested typical symptoms at the onset of HHV6-associated encephalitis. However, 11 patients presented with dysesthesia and pruritus, described as typical manifestations of patients with CIPS, and the remaining 10 showed both symptoms. Six of 11 patients with CIPS-like symptom also exhibited dysautonomia (bladder and rectal disturbance and sinus tachycardia) and/or abdominal pain. Positive results for brain MRI scans (limbic encephalitis) were observed in 9 of 14 patients (64.3%) who developed encephalitis-type symptoms, which were not found in the 11 patients who had CIPS-like symptoms; none presenting with CIPS-like symptoms had positive results for spinal MRI as well. On the other hand, 8 patients (1.8%) were diagnosed as having CIPS on the 5 to 91 days (madian 22 days) posttransplant. For graft-versus-host disease (GVHD) prophylaxis, 2 patients received cyclosporine, and 6 received tacrolimus. CI concentrations in these patients were maintained within the target ranges at onset of the pain. Abnormal findings at X-ray and MRI were not observed in all patients, but only one patient showed abnormal findings at joint of hand, finger and ankle in bone scintigraphy. For the treatment of CIPS, in 7 patients dosage of CI was reduced, whereas in one CI was switched into another one. Clinical symptoms in all of these patients were improved and exacerbation of GVHD was not seen. Conclusion: Detection of HHV6 DNA in CSF is crucial to make a differential diagnosis of HHV6 myelitis and CIPS. Transplantation physicians should be aware that CIPS-like dysesthesia and pruritus might be early manifestations to suggest the reactivation of HHV6, especially for patients who develop myelitis. Disclosures Akashi: Asahi Kasei: Research Funding, Speakers Bureau; Shionogi: Research Funding, Speakers Bureau; Astellas: Research Funding, Speakers Bureau; Celgene: Research Funding, Speakers Bureau; Chugai: Research Funding, Speakers Bureau; Bristol-Myers Squibb: Research Funding, Speakers Bureau; Novartis Pharma K.K.: Consultancy, Research Funding, Speakers Bureau; Kyowa Hakko Kirin Co., Ltd.: Consultancy, Research Funding, Speakers Bureau.

Abstract: Abstract 522 Hemophagocytic lymphohistiocytosis (HLH) is a syndrome characterized by hemophagocytosis in the bone marrow with systemic inflammatory reactions. We have reported that the inflammatory cytokinemia including interferon-gamma (IFN-gamma), tumor necrosis factor-alpha (TNF-alpha) and interleukin-6 (IL-6) might play a key role in activating macrophages to induce HLH (Akashi et al. Br J Haematol, 1994). Recent studies have shown that the engulfment of blood cells by macrophages is regulated in part by the interaction of CD47 and its ligand, signal regulatory protein alpha (SIRPA). Blocking the CD47 and SIRPA interaction by using anti-CD47 monoclonal antibodies can induce engulfment of blood cells, suggesting that the SIRPA signaling provide “don't eat me signals” to prevent macrophages from digesting self blood cells. These data led us to hypothesize that HLH is caused by the disruption of self-recognition by macrophages through impairment of the CD47-SIRPA system. To test this hypothesis, we evaluated the expression level of CD47 in bone marrow cells in 24 patients with HLH. Interestingly, the expression of CD47 was significantly downregulated (by ∼2-fold reduction in average at the protein level) in the CD34+CD38− hematopoietic stem cell (HSC) fraction in HLH patients, whereas that of the CD34+CD38+ progenitor and other mature blood cell fractions was unchanged. We then purified the CD34+CD38− HSCs and CD34+CD38+ progenitor cells from HLH patients and normal controls, co-cultured them with human macrophages in the presence of IFN-gamma and lipopolysaccharide, and evaluated the percentage of phagocyting macrophages. The numbers of phagocyting macrophages were significantly higher in cultures of CD34+CD38− HSCs from HLH patients, as compared to those in cultures of normal HSCs and of progenitor populations, suggesting that the expression level of CD47 inversely correlated with the efficiency of hemophagocytosis. Furthermore, normal HSCs but not CD34+CD38+ progenitors or other mature blood cells down-regulated CD47 in vitro in response to IFN-gamma, TNF-alpha, and IL-6, suggesting that these cytokines plays a critical role in down-regulation of CD47 especially in HSCs. In contrast, the expression level of SIRPA did not differ in myeloid cells between HLH patients and normal controls, and mutations of SIRPA were not found in HLH patients. Thus, in HLH, inflammatory cytokines down-regulate CD47 selectively in HSCs, resulting in the engulfment of HSCs by macrophages. Our data strongly suggest that the cytokine deregulation can lead to HLH in human, through disruption of self-recognition guided by the CD47-SIRPA system at the HSC stage. Disclosures: No relevant conflicts of interest to declare.

Abstract: [Introduction] Adult T-cell leukemia/lymphoma (ATL) is an aggressive peripheral T-cell lymphoma (PTCL) with a dismal prognosis. Allogeneic hematopoietic stem cell transplantation (allo-HSCT) is the only curative treatment in ATL patients. Mogamulizumab, a humanized anti-CC chemokine receptor 4 (CCR4) monoclonal antibody, is a novel immunotherapeutic agent, effective in treating patients with PTCL such as ATL, PTCL-not specified, and cutaneous T-cell lymphoma. However, in allo-HSCT setting, we should be careful to use mogamulizumab because CCR4 is expressed in regulatory T cells: The mogamulizumab treatment may accelerate GVHD by eradicating regulatory T cells in allo-HSCT patients. Here, we retrospectively analyzed the effect of mogamulizumab on GVHD development in ATL patients treated with mogamulizumab prior to allo-HSCT. [Patients and Methods] Data from the Fukuoka Bone Marrow Transplantation Group were retrospectively analyzed after the approval of mogamulizumab use in Japan. [Results] A total of 24 patients with ATL received mogamulizumab prior to allo-HSCT between April 2012 and April 2015 in our group. The median age at allo-HSCT was 58.5 years (range, 32-72). The median intervals from the last administration of mogamulizumab to allo-HSCT were 25 days (range, 9-126). The median total dose of mogamulizumab was 3 mg/kg (range, 1-8 mg/kg). After treatment with mogamulizumab, 18 patients (75%) had achieved in remission (CR in 4 patients and PR in 14) at allo-HSCT. Ten patients received unrelated bone marrow, 5 received related peripheral blood, and 9 received cord blood as stem cell sources. Eleven patients were treated with full-intensity conditioning and 13 received reduced-intensity conditioning. Graft-versus-host disease (GVHD) prophylaxis consisted of calcineurin inhibitors (cyclosporine or tacrolimus) with short-term methotrexate in 14 patients and mycophenolate mofetil in 9. The cumulative incidence (CI) of acute GVHD at 100 days was 66.6% in grade 2-4 and 33.3% in grade 3-4. The involved organs of acute GVHD were skin in 14 patients, gut in 10, and liver in 4. Among 14 patients who developed grade 2-4 acute GVHD, 5 had severe fluid retention such as pleural effusion or ascites associated with GVHD. Chronic GVHD was observed in 6 patients, and 5 of them were extensive disease. The CI of transplant-related mortality (TRM) and relapse at 1-year were 53.2% (95%CI, 29.3-72.3%) and 29.6% (95%CI, 12.6-48.9%), respectively. The leading cause of death was GVHD (n = 7). The 1-year overall survival and progression-free survival were 19.2% (95%CI, 5.7-38.8%) and 17.2% (95%CI, 4.9-35.7%), respectively. [Discussion] Use of mogamulizumab prior to transplantation in allo-HSCT patients has a merit to decrease the burden of ATL cells. However, it was associated with an increase of TRM due to severe GVHD. Although most of ATL patients achieved better disease status at allo-HSCT through mogamulizumab and the survival rate was expected to be 50% based on the previous data, the survival in the present study was ~20%. These data suggest that mogamulizumab administered before transplantation may have retained until an early phase of post-transplantation, and the donor or host-derived regulatory T cells might be eliminated, allowing the GVHD T-cell clone to expand. Since mogalizumab is a potent anti-ATL agent, we need to develop new treatment protocols integrating mogalizumab at a suitable dose or administration timing, to minimize the unwanted GVHD development in future studies. Disclosures Akashi: Asahi Kasei: Research Funding, Speakers Bureau; Shionogi: Research Funding, Speakers Bureau; Astellas: Research Funding, Speakers Bureau; Celgene: Research Funding, Speakers Bureau; Chugai: Research Funding, Speakers Bureau; Bristol-Myers Squibb: Research Funding, Speakers Bureau; Novartis Pharma K.K.: Consultancy, Research Funding, Speakers Bureau; Kyowa Hakko Kirin Co., Ltd.: Consultancy, Research Funding, Speakers Bureau.