Educational Theme | AITL: Pathophysiology, diagnostic challenges, and current and future treatment of a rare T-cell lymphoma subtype

Angioimmunoblastic T-cell lymphoma (AITL) is a subtype of mature peripheral T-cell lymphoma (PTCL) that has undergone several name changes over the years, having been called angioimmunoblastic lymphadenopathy with dysproteinemia, immunoblastic lymphadenopathy, and lymphogranulomatosis X, before finally settling into its current moniker. The name is not the only thing that has changed, however: early descriptions of AITL suggested that the entity was a nonmalignant, hyperimmune reaction or possibly a premalignant condition. Decades later, aided by diagnostic staining techniques, we now know that AITL is indeed a malignant lymphoma.¹ This rare T-cell lymphoma arises from a subset of peripheral mature CD4⁺ T follicular helper (TFH) cells and is an aggressive lymphoid malignancy that generates intense systemic inflammatory and immune reactions.^2,3

Pathobiology of AITL

Morphology

Morphologically, AITL lymph nodes demonstrate a loss of architecture due to a polymorphic lymphoid infiltrate comprised of tumor and inflammatory cells (including lymphocytes, eosinophils, macrophages, and plasma cells), as well as increased vascular arborization and extra-follicular proliferation of follicular dendritic cells (FDCs). Neoplastic cells are medium sized with clear cytoplasm and minimal atypia and are often found adjacent to the proliferative vasculature (called high endothelial venules [HEVs]). Morphologic diagnosis is challenging, as these features are often seen in association with reactive conditions, and malignant cells—in addition to being well-differentiated—are often few in number: in approximately 30% of cases, morphologically distinct clear cells are not identifiable.^3,4

There are three histologic patterns that have been described in AITL (Table 1). These patterns have some overlap, and most cases of AITL are Pattern II (occasional regression of follicles) or Pattern III (complete effacement with no residual follicles); Pattern I (largely intact nodal architecture with hyperplastic follicles) is uncommon and can be difficult to identify due to a morphology that appears more reactive than overtly neoplastic.⁵

Table 1. Histologic patterns of AITL*

Immunohistochemistry (IHC)

TFH cells in AITL are characterized by the expression of CD3, CD4, and CD10, and aberrant loss of CD5 and/or CD7; in up to 70% of cases, CD7 is not detectable.^1,3 Other variably positive TFH cell markers include CXCL13, programmed cell death-1 (PD-1), B-cell lymphoma 6 (BCL6), and inducible T-cell co-stimulator (ICOS).

According to the 2016 World Health Organization (WHO) classification guidelines for TFH lymphomas, diagnosis requires positive immunostaining for at least two, and ideally three, of the following antigens³:

• CD10
• BCL6
• PD-1
• CXCL13
• CXCR5
• ICOS
• Signaling lymphocytic activation molecule-associated protein (SAP)

AITL, like natural killer/T-cell lymphoma, has close ties to the Epstein-Barr virus (EBV), with the majority of AITL cases demonstrating active EBV infection. Interestingly, it is the reactive B cells rather than the malignant TFH cells that are EBV positive.¹ These EBV-positive B immunoblasts have the potential to develop into an EBV-positive diffuse large B-cell lymphoma (DLBCL) which can occur as a composite lymphoma or may arise at a later phase of AITL (and can further complicate diagnosis and treatment).³

Mutations and AITL development

A number of genetic mutations have been identified in relation to AITL as outlined in Table 2.

Table 2. AITL genetic mutations*

Diagnosis

Epidemiology

AITL is a rare lymphoma, comprising 1‒2% of non-Hodgkin lymphomas (NHL). It is the second most common type of PTCL (after PTCL-NOS), representing one in five of PTCL diagnoses.^1,3 The median age at diagnosis is 65 years, with men and women affected equally. Notably, unlike other PTCL subtypes, AITL is more common in Europe (28.7% of all PTCL diagnoses) than in Asia (17.9%), although this may be due to the higher incidence of NK/T-cell lymphoma in Asia.¹

Signs and symptoms

Patients with AITL often report nonspecific symptoms associated with systemic inflammation; signs are nonspecific as well, and the road to diagnosis can be a long one, spanning weeks or months. The chief complaints are fever, unintentional weight loss, and night sweats—the so-called B symptoms—and generalized non-bulky lymphadenopathy. Other findings include^1,3,5:

• Bone marrow involvement in approximately 70% of cases
• Later stage at diagnosis (Ann Arbor Stage III/IV) in 80‒90% of cases
• Rash in up to 50% of patients, with a wide range of manifestations
• Hepatosplenomegaly
• Dysproteinemia (generally a polyclonal gammopathy rather than monoclonal)
  • In some cases, circulating plasma cells may be seen on peripheral smear
• Warm autoimmune hemolytic anemia
• Polyarthritis
• Vasculitis
• Laboratory abnormalities such as elevated lactate dehydrogenase (LDH), cytopenias, and eosinophilia

The nonspecific nature of the signs and symptoms associated with AITL, combined with the difficulties associated with diagnosis based on morphology, lends to a rather lengthy list of potential diagnoses. See Table 3 for a description of other lymphoma entities that are often included in the differential diagnosis of AITL.

Table 3. Differential diagnosis of AITL*

Prognosis

Though the natural history of the disease is variable, AITL has a median survival of 5 years, with a 5-year overall survival (OS) rate of 30% and a 5-year event-free survival (EFS) rate ranging from 18% to 38%.^3,5 There is a survival advantage for patients with an international prognostic index (IPI) score of 0/1, which is associated with a 5-year OS of 56% compared with 25% for an IPI score of 4/5, although it should be noted that the IPI was developed for aggressive B-cell NHL. Attempts at a clinically-applicable prognostic index for AITL have not yet been successful.¹

AITL treatment

Based on what we have discussed so far, it is not terribly surprising that, in addition to being difficult to diagnose, AITL is also difficult to treat. Frontline treatment is similar to that used to treat other PTCLs, while further lines of treatment depend on whether or not the patient will undergo allogeneic hematopoietic stem cell transplantation (allo-HSCT).⁶

Induction regimens

Currently, there is no gold standard frontline chemotherapy regimen for patients with newly diagnosed AITL. For most patients, initial treatment is with CHOP (cyclophosphamide, doxorubicin, vincristine, and prednisone). While CHOP has relatively high complete response (CR) rates as upfront therapy in nodal PTCLs (up to 39%), its efficacy in AITL is unclear, and relapse rates are high.^1,5

Other agents have been investigated in combination with CHOP (in PTCL):

• Etoposide + CHOP (CHEOP) has demonstrated higher CR rates (51%) and may allow for deeper remissions in patients who are candidates for allo-HSCT.^1,5,6
  • In elderly patients (>60 years), the toxicity associated with the addition of etoposide negated any survival benefits over CHOP alone.⁶
• Alemtuzumab + CHOP has induced responses in a small number of patients with AITL across several phase II trials and appears to potentially improve CR compared with CHOP alone, though infectious complications are a concern.¹
• Bortezomib + CHOP demonstrated an overall response rate (ORR) of 76% and a CR rate of 65% in a phase II trial in which patients with AITL represented 17% of the study population. When compared with non-AITL populations, AITL had an improved 3-year OS advantage.¹
• Belinostat + CHOP demonstrated response rates of 86% (ORR) and 71% (CR) in a phase III trial of 23 patients with PTCL (10 of whom had AITL); nevertheless, serious adverse events (AEs) occurred in 43% of patients.⁷
• Romidepsin + CHOP (Ro-CHOP) demonstrated similar response rates, progression-free survival (PFS), and OS when compared with CHOP in a phase III study of patients with PTCL.⁸
• Brentuximab vedotin + cyclophosphamide, doxorubicin, and prednisone (BV + CHP) demonstrated superiority over CHOP in patients with CD30-positive PTCL in the phase III ECHELON-2 trial, with median PFS of 48.2 months vs 20.8 months as well as an OS benefit in patients who received BV + CHP.⁹

Non-CHOP-based regimens have also been explored as potential frontline therapy for AITL. In a retrospective study that included patients with PTCL, doxorubicin, cyclophosphamide, vindesine, bleomycin, and prednisone (ACVBP) was found to be superior to CHOP, although this regimen requires intense consolidation following the induction phase.¹ Other regimens that have thus far failed to demonstrate improvements over CHOP-based regimens include⁶:

• Etoposide, ifosfamide, and cisplatin alternating with doxorubicin, bleomycin, vinblastine, and dacarbazine
• Cisplatin, etoposide, gemcitabine, and solumedrol
• Gemcitabine, etoposide, and cisplatin

Upfront consolidation with allo-HSCT

While there is no evidence from randomized clinical trials to support the use of consolidation with allo-HSCT in first remission for patients with PTCL (including AITL), and the identification of transplant eligible patients is fraught, the use of this treatment modality has gained popularity. Patients who are chemo-sensitive by computed tomography (CT) or positron emission tomography (PET)/CT have a survival advantage over those who are not in at least partial remission at the time of transplantation.^1,6 Results from a prospective phase II study showed a 5-year OS and PFS of 51% and 44%, respectively, in patients with AITL who underwent upfront consolidation and allo-HSCT, and results from retrospective and registry-based studies are similar.⁶

Relapsed/refractory (R/R) AITL

Due to the potential for the development of DLBCL—which, again, may occur concomitantly with AITL or at relapse—repeating biopsy is necessary. Once the etiology of relapse has been confirmed, a decision must be made as to whether the goal of treatment is cure or palliative care.⁶

While the data are limited, they in general do not support the utility of autologous transplantation for R/R AITL. Allo-HSCT, however, is more promising, with reports of PFS as high as 81% in AITL. In patients with controlled disease, consolidation with allo-HSCT is an excellent option.⁶

For patients who have been selected to undergo allo-HSCT, an intensive regimen such as ifosfamide, carboplatin, and etoposide (ICE) can be used as bridging therapy, as this regimen is associated with high response rates. For patients who are ineligible for transplant, however, sustained disease control rather than a quickly induced response is the goal. Table 4 summarizes some of the agents that may be used as continuous therapy for patients with R/R AITL.⁶

Table 4. Agents for continuous therapy in R/R AITL*

Future directions

Brentuximab vedotin, an anti-CD30 antibody with an antimitotic payload, has demonstrated efficacy in the treatment of CD30-positive AITL, identifying CD30 as an important therapeutic target in this disease. Other potentially important targets in AITL include the phosphoinositide-3-kinase (PI3K) and Janus kinase (JAK)/signal transducer and activator of transcription (STAT) signaling pathways^1,6.

Case study

• Age: 71 years
• Sex: Female
• The patient had a history of generalized lymphadenopathy, and polymyalgia rheumatica that had been treated with low-dose steroids for 2 years.
• An axillary lymph node was biopsied⁵
  • On hematoxylin and eosin (H&E) stained tissue, there was a diffuse infiltrate of atypical lymphoid cells with effacement of the nodal architecture, focal follicles with regressive changes, and prominent arborizing neovascularization in the paracortex. Some Reed-Sternberg (RS)-like cells were seen
  • Flow cytometry showed an atypical T-cell population that was positive for CD2, CD4, CD5, and CD7, and negative for CD8, with loss of surface CD3. It was also observed that 45% of the T cells were positive for CD10
  • IHC confirmed a TFH phenotype, with the atypical lymphocytes expressing PD-1 and ICOS and resetting the RS-like cells, which were EBV-positive

While Hodgkin RS-like cells were present, the background of atypical lymphocytes did not support a diagnosis of classic Hodgkin lymphoma (cHL). Clonal rearrangement of TCR genes was detected via polymerase chain reaction (PCR), confirming a T-cell lymphoma diagnosis. In this case, there was a marked expansion of CD21 dendritic cells surrounding clusters of atypical lymphocytes, and this feature, combined with the presence of EBV-positive B cells, favor a diagnosis of AITL rather than nodal PTCL with TFH phenotype.⁵

Conclusion

AITL is a rare T-cell lymphoma that has long been misunderstood due to its infrequency and its variable and atypical presentations. Symptomatology is often non-specific and can mimic autoimmune diseases and other causes of systemic inflammation, and the histologic picture can add to the confusion, with the presence of RS-like cells and an atypical B-cell population. The high incidence of EBV positivity in AITL means that the development of DLBCL is a possibility, one that can heighten the uncertainty.^1-6

Decades of treating patients with AITL using protocols developed for other PTCLs has likely not served this patient population well, though advances in our understanding of this complex disease have led to the identification of therapeutic targets. Brentuximab vedotin, the anti-CD30 antibody-drug conjugate, has shown impressive efficacy in patients with CD30-positive AITL, and other agents and pathways are under investigation. Continued research regarding new targets will lead to the development of AITL-specific treatment regimens, and molecular profiling will hopefully be used to individualize therapy and improve outcomes in this patient population.^1-6