CMV seropositivity and mycosis fungoides—the Indian perspectives

I read with interest the paper by Herne et al¹  showing significant cytomegalovirus (CMV) seropositivity in patients with mycosis fungoides (MF) and Sézary syndrome (SS).

The same authors have shown previously that patients with MF/SS had significant association with HLA-DR5 and HLA-DQB ^*03 alleles.²  Of these 2 HLA antigens, HLA-DR5 was shown to be exquisitely sensitive to stimulation by staphylococcal super-antigens, which were associated with the erythrodermic phase of MF/SS.

In India, MF/SS is an extremely rare neoplasm.³  Tata Memorial Hospital in Mumbai, which is one of the biggest oncology centers in India, sees only a few patients with MF/SS out of several thousand patients with non-Hodgkin lymphoma (NHL) every year (T. Saikia, personal written communication, December 2002). Another busy tertiary care center from South India reported only 20 cases of MF/SS over 10 years, out of which 1 patient was serologically positive for human T-cell leukemia virus type 1.⁴ 

However in India, CMV seropositivity is extremely common, and in adults it reaches up to 95%.⁵  This pattern is similar to Epstein-Barr virus (EBV) seropositivity in the Indian population. In contrast to Western populations, EBV and CMV infections are acquired in childhood.⁶ 

Hence in India, high CMV seropositivity is associated with rarity of MF/SS. Is it possible that association of cutaneous T-cell lymphoma (CTCL) (MF/SS) with CMV seropositivity in the study reported depends more on late acquisition of CMV infection in the Western countries in contrast to that in India?

One argument to explain this anomaly in India may be that we in India do not have the proper HLA background for CMV infection to drive MF/SS. That is also unlikely as the HLA-DQB^*03 allele⁷  has been reported from the Indian population in association with severe lupus erythematosus, and HLA-DR5 has also been reported from this population in association with HLA-DQ3 with high haplotype frequency of 18.9%.⁸  Another argument could be that we may be missing the diagnosis in many patients in our country because routine biopsy of the various skin lesions is usually not done, and modern immunophenotyping techniques are not widely available, although very expert histopathologists with special interest in lymphoma diagnosis are available in many tertiary centers. Other conditions like alopecia areata and vitiligo, in which T-cell infiltration into the dermis may occur, are also not uncommon in India. The most common cause of vitiligo in India used to be leprosy, which is almost on the verge of eradication; hence, in the future we may get more of those subtypes of vitiligo, which may be a premycotic stage of MF/SS. However, it is unlikely that a disease such as CTCL, which is progressive in many patients, will not eventually be diagnosed en masse.

It may not be out of place to mention here that there was not a single patient of Asian origin with MF/SS in the large series of this disease presented by the authors¹  in Table 1, although several million Asians older than 30 years live in the United States (age ranges for MF/SS described in Table 1, 30-86 years). This disease, incidentally, is also rare in the whole of Asia.⁹  CMV infection might require association of some other viruses to drive CTCL in the same way as HIV 1 infection needs help from Kaposi sarcoma herpesvirus to produce Kaposi sarcoma. Hence, a search for such helper viruses in these patients is important. The prevalence of infection with different CMV genotypes may vary in different parts of the world, and this variation may also partly explain the varying pathogenic potential of this virus. From India, very few studies on CMV genotypes are available. In one such study, CMV genotypes gB subtypes II and III were found in patients with CMV disease, but no gB subtypes I and IV were found.¹⁰  As no polymerase chain reaction or in situ hybridization studies were available for patients reported,¹  the curiosity whether low levels of active CMV infection in these patients contributed to the disease remained because one of the hypotheses related to the evolution of MF/SS involves persistence of antigen (viral or others?) in the affected tissue¹¹  in a proper HLA milieu.²