Correspondence *12440 Ansin Circle Dr, Potomac, MD 20854, USA

Email dennisg@medimmune.com

The case

A 41-year-old Afro-Caribbean male presented to his local physician with a pruritic skin rash and associated skin thickening. He had been well until the rash developed 3 months previously. The patient was referred to a dermatology department where a skin biopsy was obtained; the findings were consistent with human T-lymphotropic virus type 1 (HTLV-I)-associated adult T-cell lymphoma/leukemia (ATLL). He underwent treatment with cyclophosphamide, vincristine and prednisone monthly for two courses; however, he did not return for the third course because the rash had dissipated and he could not afford additional therapy. He became lost to follow-up.

Three years later, the patient presented to his local physician with multiple problems including an 18.14 kg weight loss, skin rash, fatigue, myalgias, and swelling of the wrists. The wrist swelling had gradually worsened in the 6 months prior to seeking medical attention; the swelling initially started in the left wrist and spread to the right wrist after approximately 2 months. Although he denied having pain, the patient reported having a feeling of pressure in his wrists when pushing up from a chair. The patient was referred to the oncology service at an academic medical center, where a rheumatology consultation was requested.

Physical examination revealed diffusely scattered papular eruptions, and cervical, axillary and supraclavicular lymphadenopathy. Diffuse, non-tender, soft-tissue swelling on the dorsal aspect of both wrists was measured at 4 3 cm on the right side and 3 2 cm on the left, with slight pitting edema noted. Mild soft-tissue swelling was also observed in the proximal interphalangeal joints of the second and third fingers on the right hand and the second finger on the left hand (Figure 1A).

Figure 1 Photography and radiography of the patient's hands and wrists at referral.

(A) A photograph of the patient's hands and wrists showing diffuse, non-tender, soft-tissue swelling on the dorsal aspect of both wrists, with slight pitting noted, and mild soft-tissue swelling in the proximal interphalangeal joints of the second and third fingers on the right hand and the second finger on the left hand. (B) X-ray of the hands and wrists showing lucencies in the capitate bone of the left hand on the right side, but with no definite bony erosion.

Full figure and legend (24K)Figures & Tables indexDownload Power Point slide (98K)

The results of the patient's laboratory studies are detailed in Table 1. Tests for antinuclear antibody, rheumatoid factor, hepatitis and cytomegalovirus were negative. The test for anti-HTLV-I/II antibody was positive. Western blot analysis was reactive for rgp46-I, p53, p36, p32, p28, p26, p24, gp21, p19 and GD21, which confirmed the diagnosis of HTLV-I infection. Table 2 shows the results of peripheral blood lymphocyte phenotyping of selected antibody markers, at the time of referral and throughout his clinical course, assessed using flow cytometry. At referral, the flow cytometry results showed that of the lymphocytes in the specimen, 95% were T cells, 2.4% were B cells and 2.5% were natural killer cells. Approximately 86% of the lymphoid cells were CD2⁺, CD3⁺ (dim), CD5⁺ (dim), CD4⁺, CD52⁺, CD8^-, CD7^-, CD16^-, CD38^-, CD56^-, and CD57^-. The T cells were T-cell-receptor ⁺ and T-cell-receptor ^- (data not shown). This immunophenotype is consistent with ATLL. The B cells were polyclonal with no atypical antigen expression.

Table 1 Results of the patient's laboratory investigations at referral.

Full tableFigures & Tables indexDownload Power Point slide (109K)

Table 2 Lymphocyte phenotyping showing aberrant T-cell population consistent with the involvement of a T-cell neoplasm at the time of, and 2 months after, referral, and marked reduction in the number of affected aberrant T cells 5 months after referral.

Full tableFigures & Tables indexDownload Power Point slide (118K)

Chest X-ray findings were normal. X-rays of the hands demonstrated subtle lucencies in the capitate bone of the left hand, with no definite bony erosion (Figure 1B). Soft-tissue swelling was detected around both wrists, but was greater on the right wrist than on the left. MRI of the right wrist was performed twice: the second MRI was performed 3 months after the first. The first MRI revealed bony erosions involving the proximal carpal bones of the right hand, including the capitate. Tenosynovitis involving the extensor tendons was also observed (Figure 2A). The second MRI showed persistent extensor tenosynovitis and mild capsulitis at the wrist (Figure 2B), more prominent erosive changes at the wrist compared to the first MRI, and progressive nonarticular marrow uptake, which indicated the possibility of neoplastic marrow infiltration.

Figure 2 MRI of the patient's right wrist.

(A) A representative cut of the first MRI showing bony erosions involving the proximal carpal bones, including the capitate. (B) The second MRI performed 3 months after the first, showing persistent extensor tenosynovitis and mild capsulitis.

Full figure and legend (19K)Figures & Tables indexDownload Power Point slide (93K)

The skin biopsy performed at the time of his initial presentation was reviewed. A monotonous infiltrate of atypical small lymphoid cells showing epidermotropism and an eosinophil-rich pleomorphic intradermal cellular infiltration with focal collections of reticular cells in the presence of HTLV-I positivity was compatible with ATLL. After referral, punch biopsies were taken from the left axilla and right forearm. An atypical lymphoid infiltrate in the superficial dermis of both specimens was observed, primarily in a perivascular distribution, with the cell size ranging from medium to large. The atypical cells were CD3⁺ and CD4⁺, greater than 50% were CD25⁺, and some scattered cells were CD30⁺ and CD8⁺. The lymphocytic infiltrate extended deep into the mid dermis of the right forearm.

An excisional biopsy of a lymph node in the right axilla revealed an atypical paracortical infiltrate, with dermatopathic changes consistent with focal involvement by ATLL. A synovial aspirate and biopsy was obtained from the right wrist. At the time of the synovial aspirate, the peripheral white blood count was 13.6 10⁹ cells/l with 10% lymphocytes. Flow cytometry of the synovial fluid aspirate demonstrated that 99% of the lymphocytes were T cells, less than 1% B cells, and less than 1% natural killer cells. Approximately 89% of the lymphoid cells were CD2⁺, CD3⁺ (dim), CD5⁺, CD7^-, CD4⁺, and CD8^- T cells, consistent with HTLV-I-associated ATLL. The immunophenotypic data correlated well with the morphology of the synovium, which contained an atypical lymphoid infiltrate with cells that were strongly positive for CD3, CD4, and CD25, and negative for CD8 and CD20 (Figure 3), consistent with ATLL. The synovial tissue was evaluated for T-cell receptor -chain rearrangement, and polymerase chain reaction (PCR) of the synovial tissue was performed for HTLV-I proviral DNA. These studies demonstrated the presence of clonal rearrangement as well as HTLV-I pol sequences, but not HTLV-II pol sequences, consistent with the histologic diagnosis of ATLL.

Figure 3 Immunohistochemical analysis of synovial tissue from the biopsy specimen of the right wrist.

(A) Histological section of formalin-fixed, paraffin-embedded synovial tissue biopsy. A dense lymphocytic infiltrate composed of cells displaying nuclear atypia and mild pleomorphism is seen in the synovial tissue. The morphological appearance of this infiltrate is very suggestive of ATLL involvement. (Hematoxylin and eosin stain, 40.) (B) Immunohistochemical stain for CD3 shows the atypical lymphoid cells to be strongly positive for this T-cell marker. (CD3 immunohistochemical stain, 40.) (C) The CD3⁺ T cells are predominantly and diffusely positive for CD4 by immunohistochemistry. (CD4 immunohistochemical stain, 40.) (D) Immunohistochemical stain reveals scattered CD8⁺ T cells, which are probably reactive. (CD8 immunohistochmical stain, 40.) (E) CD25 immunohistochemical stain reveals diffuse and strong positivity in the atypical cells. This finding supports the diagnosis of ATLL. (CD25 immunohistochemical stain, 40.) (F) CD30⁺ atypical cells showing cytoplasmic and membranous staining pattern. (CD30 immunohistochemical stain, 60.) Abbreviation: ATLL, adult T-cell lymphoma/leukemia.

Full figure and legend (105K)Figures & Tables indexDownload Power Point slide (179K)

Since the patient was enrolled in an investigational drug protocol using a humanized monoclonal antibody directed against CD2, the response of his joint swelling to the treatment could be assessed as well. The investigational drug was administered twice, with a period of 2 weeks between each dose, for the treatment of his lymphoproliferative disease. Therapy was reassessed at day 28 with plans of continuing therapy if there was a favorable or stable response; however, the size of his axillary lymphadenopathy had increased and a biopsy confirmed the presence of malignant cells, indicating a lack of effective response. A PET scan performed 1 month after cessation of anti-CD2 treatment further demonstrated evidence of progressive disease. Uptake was markedly greater than normal in and around multiple joints including his shoulders, elbows, wrists, hands, and hips (see Figure 4A). There was also increased uptake noted in the neck midline, lying posteriorly at the spinous processes of a lower cervical vertebra, and probable lymph node uptake was seen bilaterally in the cervical region, the left supraclavicular region, both axillae, both inguinal regions, and epitrochlear regions. Further increased uptake was seen in the muscle between his ribs and the tips of the scapulas in the left lower back, anterior to the femoral heads. The anti-CD2 monoclonal antibody was discontinued and he was switched to another experimental protocol using alemtuzumab.

Figure 4 PET scans of the patient's wrists and hands following injection with approximately 15 mCi ¹⁸F-FDG.

(A) PET scanning prior to the initiation of alemtuzumab demonstrating marked soft tissue uptake in the hands and wrists. (B) PET scanning three months after the initiation of alemtuzumab demonstating marked improvement in the degree of uptake in the soft tissues of the wrists and hands. Abbreviation: ¹⁸F-FDG, fluorine-18 fluorodeoxyglucose.

Full figure and legend (12K)Figures & Tables indexDownload Power Point slide (86K)

Treatment with alemtuzumab, an anti-CD52 humanized monoclonal antibody, began 2 months after referral in an attempt to control the patient's overall disease process. He received seven infusions of 30 mg over a 3 month period. Restaging of his condition 2 weeks later showed overall improvement in peripheral blood lymphocyte phenotyping (Table 2) and in the extent of his adenopathy. While on this treatment protocol, his joint swelling improved considerably as reflected by the markedly reduced enhancement on MRI (not shown), as well as by PET 3 months later (Figure 4). Unfortunately, the remainder of his disease continued to smolder while on the treatment protocol. After completing the experimental protocol and returning home, an aggressive relapse of his ATLL occurred; the patient died 6 months later.

Discussion of diagnosis

HTLV-I infection is associated with a variety of human diseases, including ATLL, which is defined as a neoplastic growth of HTLV-I-infected T cells with severe organ infiltration, leading to unique clinical features depending on the organ affected.¹ HTLV-I is also associated with non-neoplastic conditions such as myelopathy/tropical spastic paraparesis, uveitis, myositis, pneumopathy and dermatitis.²

Several case reports have also described HTLV-I-associated arthropathy (HAAP), which includes synovial abnormalities that occur in individuals infected with HTLV-1.^{3, 4, 5, 6} The first report of HAAP described a patient with oligoarthritis followed by the development of a rash 1 week later.³ This patient was found to have lymphocytosis in her peripheral blood with markers characteristic of ATLL. A subsequent study reported the association between polyarthritis and HTLV-I.⁷ It was later noted that arthritis might precede the onset of ATLL by up to 5 years.⁸

The exact mechanism by which HTLV-I induces synovial proliferation is not completely understood. HTLV-I encodes tax, a transactivator/oncoprotein that exerts potent effects on synovial cells, including activation of NF-B with subsequent enhancement of cytokine expression.^{9, 10} These cytokines might have a critical role in the development of synovial hyperplasia. In addition, proteins such as HTLV-I env are recognized by expanded T-cell clones in the synovium, which also supports a possible role of HTLV-I in the development of the arthropathy.¹¹ The viral load of HTLV-I could contribute to the development of arthritis, since the circulating proviral load of HTLV-I is higher in HTLV-I-infected individuals with arthritis than in HTLV-I-infected people without arthritis. The proviral load correlates with the percentage of memory and activated CD4⁺ T cells in these patients, and a high proviral load was found in the synovial fluid and tissue cells in a patient with rheumatoid arthritis (RA).¹² These findings suggest that HAAP might result from the presentation of virus to antigen-presenting cells, with the specific proinflammatory cytokines released determined by the proviral load. Alternatively, HAAP in individuals with malignancy might also be a consequence of the infiltration of leukemic or lymphomatous cells into synovial tissues; although rare, tissue infiltration has been well documented previously.^{13, 14} It is also conceivable that both of these diagnostic possibilities can occur in the same individual since they are not mutually exclusive.

Determination of whether infiltration of the synovial tissues occurs with malignant cells alone, or whether nonmalignant cellular infiltration occurs with or without virus, might be useful in formulating an approach to the management of arthritis. In this patient, to assist in the determination of his treatment, a synovial biopsy was performed to obtain synovial tissue for diagnostic testing. Analysis of this tissue revealed an atypical lymphoid infiltrate, HTLV-I-infected cells in the synovium, and immunophenotypic staining characteristics of ATLL. T-cell receptor -chain gene rearrangement and HTLV-I sequences in the synovial tissue cells (data not shown) were detected by PCR amplification. These findings provided evidence in support of both malignant infiltration of the synovial tissue as well as the presence of HTLV-I-infected lymphocytes and synovial cells. Other investigators have detected virus in synovial cells, as well as oligoclonal proliferation to illustrate the presence of virus and malignant transformation; to our knowledge, however, none have specifically shown the improved response of the arthritis to alemtuzumab.^{15, 16, 17} The presence of HTLV-I in the synovial tissue cells is perhaps not surprising given that HTLV-I is able to enter many different cell types.¹⁸

In addition to the direct viral invasion of synovial tissue (triggering an inflammatory response) and the infiltration of synovial tissue by malignant cells, the joint swelling might have been due to reactive arthritis as a consequence of molecular mimicry, whereby a shared epitope exists between the microbe and the host joint tissues.¹⁹ Although plausible from a pathogenic viewpoint, no previous cases of HTLV-I-induced arthritis have been definitively shown to result from molecular mimicry. It has been shown, however, that molecular mimicry might exist between the HTLV-I tax protein and CNS proteins in patients with HTLV-I-associated myelopathy.²⁰ To lend credence to the hypothesis that molecular mimicry has a role in HTLV-I-induced arthritis, analogous observations to those seen in HTLV-I-associated myelopathy are needed. The presence of viral proteins in the synovial tissue that might result in the direct viral stimulation of synovial tissue cells, and the infiltration of malignant cells into the synovial tissues that was demonstrated by synovial tissue examination, implicates these viral proteins as being the more likely cause for the synovial abnormalities seen in this patient.

Treatment and management

Currently, no effective treatment is available for HTLV-I-associated ATLL; however, intense investigation is ongoing. This patient was enrolled on a clinical research protocol using alemtuzumab, a humanized monoclonal antibody directed against CD52, because of the extensive burden of malignant disease and the lack of response to other therapeutic agents. CD52 is present on most normal and malignant B and T cells, monocytes, macrophages and natural killer cells, thereby providing alemtuzumab with a broad biologic activity.²¹ While alemtuzumab has been used primarily for the treatment of B-cell chronic lymphocytic leukemia, its effectiveness has been studied in a variety of other oncologic and autoimmune disorders; several studies of alemtuzumab therapy for RA have shown substantial clinical improvement with effective amelioration of arthritis in most patients.^{22, 23, 24, 25} It is possible that a common mechanism of action of alemtuzumab exists in patients with RA and our patient, such as the elimination of synovial CD4⁺ memory T cells.

Conclusion

We have presented a patient with HTLV-I-associated ATLL complicated by arthritis, a relatively unusual manifestation in this patient population. Although disease persisted in other areas of his body while being treated on an experimental protocol using alemtuzumab, marked improvement in his arthritis was noted, indicating the possibility of differential tissue susceptibility to the effects of this potent chemotherapeutic agent. More detailed analyses of the immune system response to targeted biologic therapies will allow better understanding of the disease pathogenesis of HLTV-I-associated ATLL, and lead to the development of more effective therapeutic agents.

References

1. Hinuma Y (1985) A retrovirus associated with a human leukemia, adult T-cell leukemia. Curr Top Microbiol Immunol 115: 127–141 | PubMed | ISI | ChemPort |
2. Watanabe T (1997) HTLV-1 associated diseases. Int J Hematol 66: 257–278 | PubMed | ISI | ChemPort |
3. Haynes BF et al. (1983) Identification of human T cell leukemia virus in a Japanese patient with adult T cell leukemia and cutaneous lymphomatous vasculitis. Proc Natl Acad Sci USA 80: 2054–2058 | Article | PubMed | ChemPort |
4. Harden EA et al. (1984) Leukemia-associated arthritis: identification of leukemic cells in synovial fluid using monoclonal and polyclonal antibodies. Arthritis Rheum 27: 1306–1308 | Article | PubMed | ISI | ChemPort |
5. Sato K et al. (1991) Arthritis in patients infected with human T cell lymphotropic virus type I. Clinical and immunopathologic features. Arthritis Rheum 34: 714–721 | Article | PubMed | ISI | ChemPort |
6. McCallum RM et al. (1997) Arthritis syndromes associated with human T cell lymphotropic virus type I infection. Med Clin North Am 81: 261–276 | Article | PubMed | ISI | ChemPort |
7. Taniguchi A et al. (1988) Adult T cell leukemia presenting with proliferative synovitis. Arthritis Rheum 31: 1076–1077 | Article | PubMed | ISI | ChemPort |
8. Eguchi K et al. (1991) A case of adult T cell leukemia complicated by proliferative synovitis. J Rheumatol 18: 297–299 | PubMed | ISI | ChemPort |
9. Giam CZ and Jeang KT (2007) HTLV-I Tax and adult T-cell leukemia. Front Biosci 12: 1496–1507 | Article | PubMed | ISI | ChemPort |
10. Aono H et al. (1998) Extracellular human T cell leukemia virus type I tax protein stimulates the proliferation of human synovial cells. Arthritis Rheum 41: 1995–2003 | Article | PubMed | ISI | ChemPort |
11. Kato T et al. (2004) HTLV-I env protein acts as a major antigen in patients with HTLV-I-associated arthropathy. Clin Rheumatol 23: 400–409 | Article | PubMed | ISI |
12. Yakova M et al. (2005) Increased proviral load in HTLV-1-infected patients with rheumatoid arthritis or connective tissue disease. Retrovirology 2: 4 | Article | PubMed | ChemPort |
13. Tachibana J et al. (1993) Lymphomatous polyarthritis in patients with peripheral T-cell lymphoma. Leuk Lymphoma 11: 459–467 | PubMed | ISI | ChemPort |
14. van Soesbergen RM et al. (1982) T cell leukemia presenting as chronic polyarthritis. Arthritis Rheum 25: 87–91 | Article | PubMed | ChemPort |
15. Kitajima I et al. (1991) Detection of human T cell lymphotropic virus type I proviral DNA and its gene expression in synovial cells in chronic inflammatory arthropathy. J Clin Invest 88: 1315–1322 | PubMed | ISI | ChemPort |
16. Takahashi T et al. (2005) Treatment for HTLV-I associated arthropathy: a case study and synovial tissue culture analysis. Rheumatol Int 26: 74–79 | Article | PubMed | ISI |
17. Yamamoto K et al. (1995) Oligoclonal proliferation of human T-cell leukemia virus type I infected lymphocytes in lesions of virus-induced arthropathy. Biochem Biophys Res Commun 208: 1040–1045 | Article | PubMed | ISI | ChemPort |
18. Overbaugh J (2004) HTLV-I sweet-talks its way into cells. Nat Med 10: 20–21 | Article | PubMed | ISI | ChemPort |
19. Fujinami RS et al. (2006) Molecular mimicry, bystander activation, or viral persistence: infections and autoimmune disease. Clin Microbiol Rev 19: 80–94 | Article | PubMed | ISI | ChemPort |
20. García-Vallejo F et al. (2005) Autoimmunity and molecular mimicry in tropical spastic paraparesis/human T-lymphotropic virus-associated myelopathy. Braz J Med Biol Res 38: 241–250 | PubMed |
21. Reiff A (2005) A review of Campath in autoimmune disease: biologic therapy in the gray zone between immunosuppression and immunoablation. Hematology 10: 79–93 | Article | PubMed | ISI | ChemPort |
22. Isaacs JD et al. (1996) Campath-1H in rheumatoid arthritis—an intravenous dose-ranging study. Br J Rheumatol 35: 231–240 | Article | PubMed | ISI | ChemPort |
23. Isaacs JD et al. (1992) Humanized monoclonal antibody therapy for rheumatoid arthritis. Lancet 340: 748–752 | Article | PubMed | ISI | ChemPort |
24. Matteson EL et al. (1995) Treatment of active refractory rheumatoid arthritis with humanized monoclonal antibody Camnpath-1H administered by daily subcutaneous injection. Arthritis Rheum 38: 1187–1193 | Article | PubMed | ISI | ChemPort |
25. Weinblatt ME et al. (1995) Campath-1H, a humanized monoclonal antibody, in refractory rheumatoid arthritis. Arthritis Rheum 38: 1589–1594 | Article | PubMed | ISI | ChemPort |

Contact the journal about this article

Subject areas under which this article appears: Diagnosis, imaging and screening | Inflammation | Miscellaneous rheumatic diseases | Therapy