KIDWAI MEMORIAL INSTITUTE OF ONCOLOGY, BANGALORE Plasmablastic Lymphoma of the Gastrointestinal Tract: A rare entity with a dismal prognosis Dr. Ashok S Komaranchath, Dr. Rudresh AH, Dr. KC Lakshmaiah, Dr. Rekha V Kumar, Dr. Loknatha D, Dr. Govind Babu KIDWAI MEMORIAL INSTITUTE OF ONCOLOGY, BANGALORE Abstract Methods & Materials Discussion Discussion cont. INTRODUCTION: Plasmablastic lymphoma (PBL) is a rare and aggressive type of mature B-cell lymphoma, which is usually associated with HIV infection. The most common site of PBL is the oral cavity. Involvement of the gastrointestinal tract is rare and literature is limited to few case reports and case series. MATERIALS & METHODS: A retrospective observational study was conducted at our institute from February 2008 to January 2015 on consecutive patients presenting with plasmablastic lymphoma involving the GI tract. The presentation, clinical findings, treatment and outcome were noted. The data was compared to various case reports and series published in peer-reviewed journals. RESULTS: There were four patients diagnosed with plasmablastic lymphoma of the GI tract of which only one patient was HIV positive and was on combination anti-retroviral therapy since 2 years. In all cases, the disease was high grade with a Ki67 of more than 85%. Among the three immunocompetent patients, only one survived with therapy; however, the patient relapsed within 6 months of completion of treatment. Furthermore, the IPI score did not seem to have any prognostic significance. CONCLUSION: Plasmablastic lymphoma was seen to have a uniformly aggressive clinical course with poor outcomes even with optimal treatment. The prognosis of immunocompetent patients appears to be worse than that of HIV-AIDS patients. Although the most common histologies seen with GI lymphomas are MALT lymphomas or DLBCL, rarer and more aggressive histologies like plasmablastic lymphoma need to be kept in mind. A retrospective observational study was carried out at a referral cancer center in South India and we included patients diagnosed with plasmablastic lymphoma of the GI tract from February 2008 to January 2015. Patient medical records were reviewed for information regarding age, sex, presenting complaints, HIV status, stage, IPI score, treatment given, response to therapy and treatment outcome. Diagnosis of PBL was confirmed by histopathology and IHC studies. Complete work up of patients including haemogram, biochemistry including lactate dehydrogenase, bone marrow biopsy and cerebrospinal fluid examination, computed tomography (CT) scans of neck, chest, abdomen and pelvis was done. Stage was assigned according to the Cotswold modification of the Ann Arbor staging system. The Eastern Cooperative Oncology Group scale was used to determine performance status. Response to treatment was determined as per the International Working Group Criteria. Chemotherapy, radiotherapy or a combination of both have been tried in plasmablastic lymphoma. The most common chemotherapy protocols used in plasmablastic lymphoma is CHOP or CHOP like regimen. More aggressive regimens have not shown to produce a statistically significant improvement in outcome. In our series, both patients with stage I disease were started with CHOP chemotherapy. Rituximab was added in case 3 as her immunohistochemistry showed CD20 positivity. However, both patients progressed on treatment. Case 1 has been started on a salvage regimen of CODOX-M/IVAC. The two cases with stage IV disease had a very aggressive course and chemotherapy could not be started. Plasmablastic lymphoma (PBL) is a neoplasm of the mature B cells which shows diffuse proliferation of the large neoplastic cells, most of which resemble B‑immunoblasts and have the immunophenotype of plasma cells.[2] Though PBL was originally described in the oral cavity of HIV-positive patients, this disease can be associated with other types of immunodeficiency. HIV positive patients with PBL are almost always found to have a concurrent EBV infection. PBL has been less frequently described in immunocompetent patients and in extra-oral locations. Based on immunohistochemical, molecular and genetic studies, the cell of origin is thought to be a blastic proliferative B cell that has switched its phenotype to a plasma cell stage of differentiation or is probably in transition from immunoblast to plasma cell. There is a strong association seen with EBV especially in the setting of HIV infection. Plasmablastic lymphoma has a characteristic immunophenotype, wherein they are positive for the plasma cell markers such as CD38, CD138 and MUM1; and negative for the typical B‑cell antigens like CD20 and CD79a [1]. There is usually a high level of proliferation as detected by high Ki-67 levels [1]. Ki-67 values more than 80% has been seen to be an independent poor prognostic factor in various studies.[11][23][24] All four patients in our series had a Ki-67 over 80% and only one patient was alive 6 months after primary diagnosis. LMP1 is done in all cases of HIV associated lymphomas. In the sole HIV positive patient in our series (Case 4) LMP1 was found to be focally positive. In a previously published series of gastrointestinal plasmablastic lymphoma, the most common location appeared to be the distal gut with two out of the four cases arising from the small intestine, one from the colon and one from the anorectum.[3] However, two of these patients had history of long standing Crohn’s disease. There was only one patient in that series who was HIV positive and he was diagnosed with disease involving the anorectum. In our series as well, the only HIV positive patient had involvement of the anorectum. A head to head comparison of the two series’ is given in Table 2. The Ann arbor staging does not appear prognostic and stage I disease should be treated the same as those with advanced disease. Patient Characteristics Results There were four patients diagnosed with gastrointestinal plasmablastic lymphoma in the time period mentioned. There were three males and one female (median age 29.5 years; range 13-45 years). [Patient details - Table 1]  Abdominal pain was the most common symptom and was present in all 4 cases. Two of the cases had stage I disease and the other two was stage IV at presentation. Both patients with stage IV disease also had ‘B’ symptoms. One of these was a 13 year old patient who also had spleen and marrow involvement. This patient had a rapidly deteriorating illness and died within 2 weeks of presentation before chemotherapy could be instituted. The other patient with stage IV disease was a HIV positive patient who was on treatment with HAART 3 years prior. Chemotherapy was deferred in in view of a poor PS and after discussion with the patient and his relatives; he was given best supportive care. He died due to disease related complications within 3 months of diagnosis. Histopathology showed a plasmablastic picture with a mixture of plasmablasts and immunoblasts. The immunohistochemistry was positive for plasma cell markers, CD138 [Fig.2] and MUM1 [Fig.3]; but negative for CK, CD3, CD20 and CD117 as well as for PAX5 and LCA. LMP1, which is a surrogate marker for Epstein Barr Virus infection, was negative for all cases except the HIV positive patient for whom it was focally positive. Ki-67 index was at least 85% or above in all cases.[Fig.4] Conclusion Introduction Plasmablastic lymphoma is a rare entity seen most commonly in the setting of immunocompromise. The most common site involved in these patients is the oral cavity. However, this disease may present in extra-oral sites in which the gastrointestinal tract appears to be the most common. GI involvement also appears to be more common in immunocompetent patients. Neither the clinical stage, nor the IPI score seems to have any prognostic significance. The prognosis is uniformly poor even with optimal management with cytotoxic agents. Plasmablastic lymphoma (PBL) is a clinico-pathological entity that was initially described in 19971 and is now considered a distinct subtype of diffuse large B-cell lymphoma (DLBCL) seen more commonly in patients with HIV infection. Involvement of the GI tract is rare and there has been only one case series of four cases3 and several case reports4-10 published in recent times. The diagnosis of PBL is challenging because it has features that overlap with those of myeloma and with lymphomas that have plasmablastic morphology.16 This is compounded by its aggressive clinical course with frequent relapses, high rates of disease progression and mortality despite the use of optimal treatment modalities.17 We hereby present our experience with plasmablastic lymphoma involving the GI tract with the clinical presentation, investigations and treatment outcomes. Table 2: Comparison of present series with Luria et.al. Figure 1 Figure 2 Figure 3 Figure 4