Adult T-Cell Leukemia/Lymphoma

Abstract

Adult T-cell lymphoma/leukemia (ATL) is a rare T-cell lymphoproliferative neoplasm caused by human T-lymphotrophic virus 1. In its more common, aggressive forms, ATL carries one of the poorest prognoses of the non-Hodgkin lymphomas. The disease has clinical subtypes (ie, acute, lymphoma, chronic, and smoldering forms) defined by the presenting features, and therefore, the clinical course can vary. For the smoldering and lower-risk chronic forms, combinations involving antiviral therapies have shown some success. However, in many patients, the more indolent forms will evolve into the more aggressive subtypes. In the more aggressive acute, lymphoma, and higher-risk chronic forms, the literature supports initial treatment with combination chemotherapy followed by allogeneic transplantation as a potentially curative approach. Recently, mogamulizumab and lenalidomide have shown promise in the treatment of ATL. With better understanding of the molecular drivers of this disease, we hope that the therapeutic landscape will continue to expand.

INTRODUCTION

Adult T-cell lymphoma/leukemia (ATL) is a rare lymphoproliferative neoplasm of mature CD4+ CD25+ T cells caused by infection with the retrovirus human T-lymphotropic virus type 1 (HTLV-1). The aggressive subtypes of ATL (ie, acute, lymphoma, and poor-risk chronic) carry some of the poorest prognoses of any of the non-Hodgkin lymphomas. In a large retrospective analysis, patients with ATL had 5-year failure-free and overall survival (OS) of only 12% and 14%, respectively.¹ However, the clinical course can be quite varied with the chronic and smoldering variants of this disease.

EPIDEMIOLOGY AND PATHOGENESIS

In the United States, the incidence of ATL is approximately 0.05 per 100,000. However, in regions where HTLV-1 is endemic (eg, regions of Japan), the incidence has been reported to be as high as 27 per 100,000.² Regions with the highest incidence of HTLV-1 include the southern and northern islands of Japan, the Caribbean, Central and South America, intertropical Africa, Romania, and northern Iran. The International Peripheral T-Cell Lymphoma Project has reported that ATL constitutes 25% of cases of T-cell lymphoma in Japan compared with 1% to 2% in Europe and North America.^1,3-5 The mean age of patients with ATL is 62 years, without a sex predominance.

In the United States, approximately 0.4% of volunteer blood donors are infected with HTLV-1 or -2,⁶ which are retroviruses with 60% nucleotide similarity.⁷ HTLV-1 has infected approximately 10% to 15% of the population in many parts of Central and South America, southern Japan, Africa, and the Middle East.^8,9 Unlike HTLV-1, HTLV-2 is not clearly associated with disease.¹⁰

The HTLV-1 genome includes Gag, Pol, and Env genes, as well as several regulatory genes, including those encoding the transcriptional transactivator protein Tax and the helix-basic-loop zipper protein HBZ.¹¹ Tax is critical for immortalization of T cells but is highly immunogenic, and thus, its expression is repressed when ATL is clinically apparent.¹¹ In cases of ATL, the HTLV-1 provirus has acquired mutations that prevent expression of viral proteins.¹² Alternatively, the promoter regulating expression of Tax and viral proteins may be deleted or hypermethylated.^13,14 In contrast, HBZ has weak tumor initiator activity; it is thought to function primarily in tumor maintenance, and it is continuously expressed in all ATL cases.¹⁵ The number of infected cells within an individual changes as a result of virus replication, release, and spread via cell-to-cell transmission¹⁶; infected cell division and resistance to apoptosis¹⁷; and cytotoxic T-lymphocyte (CTL) clearance of cells expressing viral proteins.¹¹ Thus, therapies directed exclusively at steps in the virus replication cycle may have limited effects on the total number of infected cells. ATL patient cases carry a single provirus integrated almost randomly, but a majority of cases represent a clonal outgrowth of cells with a single integration site.¹⁸ The histone deacetylase inhibitor valproate activates Tax and Gag expression, enhances CTL activity, and results in reduced proviral load.^19,20 Other recent data suggest that HTLV-1–specific CTLs can be induced in patients with ATL and that these CTLs can contribute to treatment and inhibit relapse. This provides important evidence supporting the concept that immunotherapy can be effective in this disorder.

Among approximately 10 to 20 million HTLV-1 carriers, the lifetime risk of developing ATL is approximately 2.5% to 4%, with a mean latency of more than 50 years.^4,21 HTLV-1 is transmitted through breastfeeding, blood products, and unprotected sexual intercourse. The overwhelming majority of ATL cases occur in patients infected during the early years of life.⁴ Of note, HTLV-1 has also been associated with HTLV-1–associated myelopathy-tropical spastic paraparesis, which is a chronic inflammatory disease affecting the CNS. Patients present with progressive spastic paraparesis, lower-limb sensory disturbance, and bladder or bowel dysfunction.²²

PATHOLOGY

First described in 1977,²³ the malignant cells carry a flower-like or clover-shaped appearance characterized by condensed nuclear chromatin and inconspicuous nucleoli. At times, these can be difficult to distinguish from Sezary cells (Fig 1). At least 5% of circulating abnormal T lymphocytes are required to diagnose ATL in patients without histologically proven tumor lesions.²⁴ These cells express the surface T-cell lymphocytic markers CD2, CD4, CD5, CD45RO, CD29, and T-cell receptor (TCR) αβ and are usually negative for CD7, CD8, and CD26 and show reduced CD3 expression. The lymphocytic activation markers HLA-DP, DQ, DR, and interleukin-2Rα (CD25) are always present, whereas terminal deoxynucleotidyl transferase is typically absent.²⁵ Rare immunophenotypic variants (CD4-negative, CD8-positive, and double-positive or double-negative variants) have been reported.^26-28 The minimal flow cytometric analysis should include CD3, CD4, CD7, CD8, and CD25. Clonal rearrangements of the TCR genes are typically present.^29-31 In addition, ATL can be associated with several histologic subtypes, including diffuse, poorly differentiated small-cell lymphoma; mixed large- and small-cell lymphoma; and large-cell immunoblastic lymphoma.³²

Fig 1.

(A) Lymph node biopsy and (B) peripheral smear in adult T-cell lymphoma/leukemia (ATL). The neoplastic cells in ATL characteristically have a flower-like or clover-shaped appearance. By flow cytometry, they express CD2, CD4, CD5, CD45RO, CD29, and T-cell receptor αβ and do not express CD7, CD8, or CD26. CD3 expression is often reduced. Photographs courtesy of Ahmet Dogan.

The genomic landscape of ATL is being elucidated, with a vast majority of ATL patient cases carrying mutations in the TCR/nuclear factor kappa B activation pathway. Sequence analysis of ATL cells demonstrates a high rate of somatic nonsynonomous mutations, significantly higher than that in most other hematopoietic malignancies.¹² The somatic mutations occur frequently in proteins that form the Tax interactome, thus providing a mechanism to substitute for Tax, which is repressed after tumor initiation. A high rate of mutations was found in genes encoding mediators of the TCR pathway and subsequent nuclear factor kappa B activation activation, analogous to that described in the B-cell receptor pathway in B-cell lymphomas. Thus, therapies targeting the TCR pathway have a role in ATL treatment. Also detected were frequent deletions, mutations, or hypermethylation of genes encoding components of the class I major histocompatibility complex, death receptors, and proteins involved in cell adhesion or immune checkpoints.¹² In particular, one subgroup contains mutations in IRF4, which is implicated in the mechanism of action of lenalidomide.¹² Whole-genome sequencing efforts have also demonstrated that ATL exhibits both gain-of-function and loss-of-function mutations in RHOA.³³

CLINICAL FEATURES

The disease is typically subdivided into four clinical presentations: smoldering (generally indolent), chronic (variable course), lymphoma, and acute (both aggressive), with the poorest prognosis seen among those with lymphoma or acute as well as poor-risk or relapsed chronic subtypes.²⁴ The acute variant of ATL represents 60% of cases and is characterized by a leukemic presentation with or without lymphadenopathy and/or visceral disease. An additional 20% of patients present with the lymphoma variant, which is characterized by lymphadenopathy and an absence of leukemic features (ie, < 1% of leukemic cells in the peripheral blood). A majority of patients present with hepatosplenomegaly (50% of cases), lymphadenopathy (almost all cases), elevated lactate dehydrogenase, hypercalcemia (50% of cases), and visceral and cutaneous lesions. The bone marrow is involved in approximately 35% of cases.³² Additional extranodal sites of disease include the lung, liver, skin, GI tract, and CNS, in which the disease can manifest itself as cord myelopathy or spastic paraparesis.

In contrast to the acute and lymphoma forms, the smoldering form of ATL typically presents with cutaneous or pulmonary lesions without visceral or bone marrow involvement and with a low level of peripheral blood (< 5% of lymphocytes) involvement. Chronic ATL presents with leukocytosis, lymphadenopathy, and organomegaly without elevated lactate dehydrogenase or visceral involvement.

Overall, prognosis in ATL remains poor. For those with the acute or lymphoma subtype, the median survival is less than 1 year. Although smoldering and chronic ATLs are characterized by indolent courses initially, the prognosis remains guarded, with 5-year survival of 40% and 50%, respectively, and approximately half of these patients experience progression to acute ATL.³⁴

Opportunistic infections are common in patients with ATL, even indolent forms.³⁵ Pneumocystis carinii infections, strongyloides, and cryptococcal meningitis are common, as are bacterial and other fungal infections.

TREATMENT

Despite limited long-term efficacy, cytotoxic combination chemotherapy remains the mainstay of therapy for ATL. Prospective clinical trials from Japan using regimens such as CHOP (cyclophosphamide, doxorubicin, vincristine, and prednisone) or an aggressive combination chemotherapy, VCAP-AMP-VECP (vincristine, cyclophosphamide, doxorubicin and prednisone; doxorubicin, ranimustine, and prednisone; and vindesine, etoposide, carboplatin, and prednisone), for patients with the chronic, lymphoma, and acute subtypes showed median OS rates from 8.3 to 10.6 months.^36,37 These initial therapies provide response rates of approximately 70%, with a complete response rate of approximately 30%.³⁷ In a phase III trial comparing VCAP-AMP-VECP with CHOP-14, CHOP-14 carried a 25% complete response rate and a 13% 3-year OS, whereas VCAP-AMP-VECP was statistically superior, with a 40% complete response rate and 24% 3-year OS.³⁸ Both arms in these studies included CNS prophylaxis with intrathecal therapy. Dose-adjusted EPOCH (etoposide, prednisone, vincristine, cyclophosphamide, and doxorubicin) has also been studied in ATL. Using dose-adjusted EPOCH with maintenance interferon and zidovudine, a study of 19 patients showed an overall response rate of 58%, with two complete remissions (CRs); however, the maintenance therapy did not seem to add to the efficacy.³⁹ A more recent study of 18 patients treated with dose-adjusted EPOCH in combination with raltegravir and bortezomib showed an overall response rate of 67%, with three CRs and eight partial remissions.⁴⁰

Unfortunately, despite aggressive first-line approaches, nearly 90% of patients experience relapse (often within months of completing therapy), and currently, consolidation of first remission with allogeneic stem-cell transplantation is strongly considered.³² In the largest retrospective series regarding the role of allogeneic transplantation in ATL, 586 Japanese transplant recipients had a 3-year OS of 36%, with similar outcomes between myeloablative approaches and reduced-intensity transplantation approaches.⁴¹ Retrospective series have shown that donor HTLV-I seropositivity adversely affected disease-associated mortality, and given the typical acquisition of the virus from the mother, it is often difficult to find matched siblings unaffected with the virus.⁴² Autologous stem-cell transplantation does not seem to provide benefit in managing ATL.^43,44 For patients with relapsed or refractory ATL, the prognosis is even poorer.

In terms of alternate approaches, the role of antiviral therapy in ATL remains controversial. Although smaller initial studies regarding the use of antiviral therapy in combination with interferon were quite promising, studies in the United States have shown less promising results.^45-47 Zidovudine remains the most studied antiviral used in this setting; however, other studies have incorporated raltegrevir and lamivudine.^39,40 A meta-analysis of 254 patients with ATL demonstrated that for patients with acute, chronic, or smoldering ATL, there seems to be a benefit to first-line zidovudine including interferon, whereas patients with the lymphoma variant did not benefit.⁵ Patients with chronic or smoldering ATL treated with first-line antiviral therapy had a 100% 5-year OS. Patients received antiviral therapy in combination with either chemotherapy or an interferon. In this meta-analysis, patients with acute ATL treated with chemotherapy with or without antiviral therapy in the first-line setting had a 5-year OS of 28% compared with 10%, respectively. Additionally, maintenance antiviral therapy in patients treated with first-line chemotherapy also conferred an improved OS.⁵ However, when evaluated prospectively, the addition of antiviral therapy to initial therapy or as maintenance after combination chemotherapy has not yielded improved results.^39,40

In 2015, mogamulizimab, an anti-CCR4 antibody, was approved in Japan for relapsed or refractory ATL. In a multicenter phase II study of 28 patients with relapsed or refractory ATL, treatment with mogamulizumab resulted in a 50% response rate and median OS of 13.7 months.⁴⁸ A randomized phase II study of modified VCAP-AMP-VECP with or without mogmulizumab showed that the combination had a superior CR rate of 52%, compared with 33% with VCAP-AMP-VECP alone, at the cost of some increases in toxicity.^49,50 A subsequent international study (in North America, Europe, the Caribbean, and South America) randomized phase II study of mogamulizumab in relapsed acute, lymphoma, or chronic ATL versus investigator’s choice of pralatrexate, gemcitabine plus oxaliplatin, or DHAP (dexamethasone, cisplatin, and cytarabine) showed an overall response rate of 28% with mogamulizumab compared with only 8% in the investigator’s choice arm.⁵¹

In addition to poor prognosis, patients with ATL have either been underrepresented or excluded from many trials of novel agents recently tested in T-cell lymphoma. Those with ATL were excluded from the registration studies of romidepsin, belinostat, and alisertib. In the phase II Southwest Oncology Group trial of alisertib, four patients with ATL were treated, and one of them responded.⁵² In the development of pralatrexate, there were a total of three patients included on the pivotal phase I and II trials. There has otherwise been only anecdotal use reported with brentuximab vedotin and romidepsin.^53,54 Most recently, a phase II study from Japan of single-agent lenalidomide showed a 42% overall response rate in relapsed or refractory ATL. Interestingly, although the median progression-free survival with this approach was 4 months, the median OS was 20 months.⁵⁵ Checkpoint inhibitors such as nivolumab are currently being studied in ATL as well in the United States (ClinicalTrials.gov identifier: NCT02631746).

OUR APPROACH

As reviewed in this article, our current results in treating ATL have been poor. We therefore strongly encourage participation in clinical trials investigating new approaches to this disease. In the absence of a clinical trial, for those with acute, lymphoma, or poor-risk chronic subtype, our preference is to use EPOCH-based regimens. Our results with EPOCH mirror those seen with VCAP-AMP-VECP in terms of response (unpublished data; Memorial Sloan Kettering Cancer Center). We try to consolidate remissions (when achieved) with allogeneic stem-cell transplantation and therefore refer for transplantation consults early in the treatment course. Because our goal of initial therapy is remission adequate to allow allogeneic stem-cell transplantation, we believe achieving a CR before a full course of EPOCH is adequate. For example, those in CR after four cycles who have ready donors may proceed to consolidation before completing a full six cycles. When the only available donors are also HTLV-1 positive, we will still proceed with transplantation, because despite the poorer results, this remains the only potentially curative strategy for these patients. In our experience, nearly 30% of our patients with ATL will have evidence of CNS involvement at diagnosis or relapse.⁵⁶ Although it is not clear whether CNS prophylaxis with intrathecal therapy has reduced this rate, we routinely include it as part of initial therapy when administered with curative intent. For patients with relapsed or refractory disease, we recommend a clinical trial whenever possible. When a clinical trial is not available or patients decline, we use one of the agents we have described here with at least anecdotal evidence of response (mogamulizumab is not currently available outside of Japan) or extrapolate from other drugs or regimens shown to be active for relapsed peripheral T-cell lymphomas.

In conclusion, ATL remains a rare T-cell lymphoma caused by infection with HTLV-1. Despite efforts to intensify therapy with multiagent chemotherapeutic approaches, the prognosis remains poor, especially for those with the acute, lymphoma, or poor-risk chronic subtype of the disease. Recent studies have shown promising activity of nonchemotherapeutic agents such as mogamulizumab and lenalidomide, and it is hoped that these could form important parts of more-effective combination regimens.

AUTHOR CONTRIBUTIONS

Conception and design: All authors

Manuscript writing: All authors

Final approval of manuscript: All authors

Accountable for all aspects of the work: All authors

AUTHORS' DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST

Adult T-Cell Leukemia/Lymphoma

The following represents disclosure information provided by authors of this manuscript. All relationships are considered compensated. Relationships are self-held unless noted. I = Immediate Family Member, Inst = My Institution. Relationships may not relate to the subject matter of this manuscript. For more information about ASCO's conflict of interest policy, please refer to www.asco.org/rwc or ascopubs.org/journal/jop/site/misc/ifc.xhtml.

Neha Mehta-Shah

Research Funding: Bristol-Myers Squibb, Celgene, Verastem

Lee Ratner

No relationship to disclose

Steven M. Horwitz

Consulting or Advisory Role: Celgene, Millennium Pharmaceuticals, Kyoto-Hakka-Kirin, Seattle Genetics, Forty Seven, HUYA Bioscience International, Infinity Pharmaceuticals

Research Funding: Celgene, Infinity Pharmaceuticals, Seattle Genetics, Spectrum Pharmaceuticals, Takeda Pharmaceuticals, Kyowa Hakko Kirin, ADCT Therapeutics, Forty Seven, Aileron Therapeutics