Primary pulmonary lymphoma (PPL) is defined as an extranodal lymphoma that only arises from the lung parenchyma or bronchus . PPL is rare, accounting for 3–4% of all extranodal lymphomas . Most cases reported in the lungs are B-cell non-Hodgkin’s lymphoma (NHL) . Non-B cell lymphomas, including T-cell and NK cell lymphomas involving the lung parenchyma, are seldom reported. Tamura and co-workers reported 24 cases of primary pulmonary lymphoma, only one of which had originated in the T-cells . The PubMed database was searched going back to 1990, and only 16 cases of T-cell lymphoma were found . Primary lung NK/T cell lymphoma is more rare, and only four cases have been reported in the literature [6–8].
Clinically, patients with primary lung NK/T cell lymphoma usually present with cough, fever, and dyspnoea. Treatment with antibiotics does not improve fever. Pleural effusion has also been reported. The most common radiographic findings are bilateral diffuse nodular lesions with mass-like consolidation. Histopathologically, nasal-type NK/T cell lymphoma has a characteristic histologic pattern, which is often angiocentric with prominent necrosis and vascular destruction. For this reason, most NK/T lymphomas may show extensive necrosis and are thus easily misdiagnosed as infectious lesions. Immunohistochemically, the tumor cells tend to express CD56, cytoplasmic CD3ϵ, CD2, cytotoxic granule proteins, granzyme B, and TIA-1 but not surface CD3. Of course, some special cases, such as CD20-positive NK/T-cell lymphoma, have been reported [14–16]. In this case, the tumor cells were negative for CD20. Thus, a diagnosis of NK/T cell lymphoma was first considered based on the histopathological characteristics and typical immunohistochemical features outlined above. Moreover, there were no skin eruptions, and no evidence of lymphoma involvement was found in the nasal endoscopy and positron emission tomographic (PET) imaging. A bone marrow biopsy was also negative. In this way, a diagnosis of primary pulmonary extranodal NK/T cell lymphoma, nasal type was made.
Extranodal NK/T cell lymphoma is an uncommon subtype of NHL. It is derived from either activated NK cells or, rarely, cytotoxic T-cells . If a patient with NK or T-cell tumors has unusual reactions to treatment or unusual prognosis , it is better to differentiate the NK tumors from the T-cell tumors in our opinion. Simultaneously, several studies have demonstrated that it is necessary to elucidate the origin of NK/T cell lymphoma [18, 19]. Unfortunately, related research is still very limited. Thus, our issue is to determine whether the tumor originated from NK or T cells.
Monoclonality is one of the main features of most tumors, but normal and reactive hyperplastic lesions are polyclonal . Clonal analysis techniques, including T-cell and B-cell receptor gene rearrangement and one technique based on X-chromosome inactivation mosaicism and polymorphisms at the PGK and AR loci in female somatic cells, play an important role in differentiating neoplasm from reactive hyperplasia. In particular, T-cell receptor (TCR) gene rearrangement may be used to distinguish NK/T-cell lymphoma from T-cell lymphoma. If TCR gene rearrangement shows a clonal pattern, the tumor is of T-cell origin. If negative, a diagnosis of NK/T cell lymphoma can be made . The point has been confirmed by many NK/T cell lymphoma reported in the literature . That is to say, determination of TCR gene rearrangement by PCR-based analysis may not be useful for making diagnosis of extranodal NK/T cell lymphoma. In the present study, DNA was isolated from paraffin-embedded tissue samples and analyzed for TCR gene rearrangement using PCR-based techniques. The results showed no evidence of clonal gene rearrangement. As pointed out by Mansoor et al., this condition seemed to be diagnosed as NK-cell lymphoproliferative lesions, specifically polyclonal lesions . However, the truth of the matter was quite different.
Clonality assays, which are based on X-chromosome inactivation mosaicism and polymorphisms at the PGK and AR loci in female somatic cells, are a very important means of differentiating neoplasm from reactive hyperplasia. The principle behind this assay is that each female somatic cell contains two X chromosomes, one of which is randomly inactivated by methylation during early embryogenesis and the other of which retains its genetic activity throughout life. PGK gene polymorphism shows the presence of a single nucleotide polymorphic site identified by Bst XI located downstream of the methylation site. AR polymorphism shows different lengths of CAG short-tandem repeats (STR) at exon 1 . After digestion with the methylation-sensitive restriction enzyme Hha I, normal and reactive hyperplastic tissues with polymorphisms show two alleles of equal intensity. Neoplastic tissues show only one of the two alleles, with obviously reduced intensity. This lesion should be diagnosed as true NK cell lymphoma or T cell lymphoma if it shows a monoclonal pattern based on the histopathological features, immunohistochemical characteristics, and TCR. In the present study, the clonal status of the lesion was evaluated using laser microdissection and a clonal assay to further confirm the conclusion. The results demonstrated that the lesion was monoclonal hyperplasia, which supported the diagnosis of true NK-cell lymphoma. Six of the eight patients described by Mansoor et al. were female . For this reason, clonality assays are here recommended as a means of confirming the nature of these lesions. Moreover, TCR gene rearrangement was not performed in any of the four cases reported in the literature [6–8]. This suggests that the diagnosis of NK cell lymphoma of the lung was not confirmed. Of course, this method has its limitations. Specifically, it can only be used to analyze samples from female individuals. Thus, the diagnosis of most patients with NK/T cell lymphoma still depends upon morphology, immunophenotype, EBER in situ hybridization, and TCR gene rearrangement.
Remarkably, there are many similarities between extranodal NK/T cell lymphoma with advanced stage and aggressive natural killer cell leukemia/lymphoma, such as the morphology, immunophenotype, germline configuration of TCR gene and EBV association . However, Kwong et al.  has pointed out the latter can be different from the former by the absence of a previous history, a shorter illness, a younger age of presentation and an extremely aggressive course. In our case, the blood count was normal, and the tumor cells did not involve in bone marrow, liver, and spleen. Moreover, the patient was an older female except showing a short survival time. Thus, we considered that it should be diagnosed as extranodal NK/T cell lymphoma.
The prognosis of primary pulmonary NK/T-cell lymphoma is very poor, although aggressive treatments with CHOP-based chemotherapy and surgical resection have been reported in the literature. Among the four patients with primary pulmonary NK/T cell lymphoma, the longest life expectancy was less than 6 months . Recently, some studies demonstrated that the expression of both LMP1 and LMP2A showed significant correlations with the prognosis of patients with extranodal NK/T cell lymphoma . In our case, the tumor cells were positive for LMP1. Therefore, the patient died 1 month after leaving the hospital, which supported the conclusion. Moreover, it appears to be very difficult to diagnose NK/T cell lymphoma because of the associated extensive necrosis and variable morphology. In the present case, frozen sections were examined twice during the operation, and the results showed microscopically extensive coagulative necrosis. Fortunately, the patient underwent a lobectomy and consented to allow pathologists to examine a large number of sections.
In conclusion, this is the first reported case of genuine primary pulmonary NK cell lymphoma. Pathologists should carefully investigate the whole lesion when extensive coagulative necrosis with atypical lymphocyte infiltration is observed in a major lesion. This renders primary pulmonary NK cell lymphoma less likely to be misdiagnosed as an infectious lesion. In addition, TCR gene rearrangement and clonal assays based on the polymorphisms of PGK and AR and X-chromosomal inactivation mosaicism should be performed in any somatic tissue samples taken from female patients.