Tomás Álvaro

PURPOSE: Recent studies of Hodgkin's lymphoma (HL) have suggested that the presence of regulatory T cells in the reactive background may explain the inhibition of the antitumoral host immune response observed in these patients. This study aimed to assess the relevance of regulatory T cells and CTLs present in the background of HL samples in the prognosis of a series of classic HL (cHL) patients. EXPERIMENTAL DESIGN: Expression of granzyme B and TIA-1 (markers for CTL) and FOXP3 (a marker for regulatory T cells) were evaluated independently by immunohistochemistry in tissue microarrays of 257 cHL patients and correlated with patient outcome. RESULTS: The combined influence of the presence of FOXP3(+) and TIA-1(+) cells distinguished three risk groups of patients with 5-year overall survival of 100%, 88%, and 73%. The presence of a small number of FOXP3(+) cells and a high proportion of TIA-1(+) cells in the infiltrate represent an independent prognostic factor that negatively influenced event-free survival and disease-free survival in cHL. Compared with the features at diagnosis, relapsed samples tended to have more TIA-1(+) cells and a lower proportion of FOXP3(+) cells in the reactive background. CONCLUSIONS: These data suggest that low infiltration of FOXP3(+) cells in conjunction with high infiltration of TIA-1(+) cells in cHL may represent biological markers predicting an unfavorable outcome. Moreover, the variation of these markers over the course of the disease implies a possible role for them in the progression of HL cases.

PURPOSE: Recent molecular data have suggested that non-neoplastic cells are powerful modulators that may confer a selective advantage or disadvantage on the outcome of follicular lymphoma (FL) patients. PATIENTS AND METHODS: The prevalence of the principal inflammatory and immune-infiltrated cells was measured immunohistochemically in the tissue of 211 FL patients, and associations were sought with their traditional clinicobiologic characteristics. RESULTS: Our results confirmed the presence of a large number of T lymphocytes (CD4+ and CD8+) and CD57+ cells and, at a moderate level, the presence of TIA-1+ cytotoxic cells, CD68+ macrophages, CD123+ plasmacytoid cells, and FOXP3+ regulatory T cells among the pool of non-neoplastic cells. In addition to the conventional clinical variables, univariate analysis identified a low level of infiltrated CD8+ T lymphocytes as a significantly negative prognostic factor of overall survival. The following significant differences in the abundance of cells of specific and nonspecific immunity were observed in relation to the clinicobiologic features of FL: (1) a reactive microenvironment mainly made up of T lymphocytes and macrophages was significantly associated with a favorable clinical behavior of FL patients; and (2) a reactive microenvironment infiltrated predominantly by CD57+ T cells was associated with a significantly higher frequency of adverse clinicobiologic manifestations such as "B" symptoms and bone marrow involvement. CONCLUSION: Our results demonstrate the existence of two specific patterns in the reactive microenvironment of FL, an immunosurveillance pattern (T lymphocytes and macrophages) and an immune-escape pattern (CD57+ T cells), that were directly associated with the clinicobiologic features of these patients.

The clinical, morphological, and immunohistochemical findings in six cases in six cases of ovarian endometrioid tumors (five endometrioid carcinomas and one carcinosarcoma) with a yolk sac tumor (YST) component are described. The age of the patients ranged from 31 to 73 years (average, 53), and only two patients were premenopausal. Two cases were stage Ia tumors, three stage III, and one stage IV. A substantial postoperative elevation of alpha-fetoprotein (AFP) was seen in two patients and a mild increase in another two. All six patients had surgery and postoperative cisplatin-based chemotherapy regimens, four of whom died of tumor 3 to 14 months after surgery without response to treatment. Only a stage Ia patient is alive and well 1 year after surgery. The tumors were large (average, 17 cm). Benign endometrioid lesions were found in the homolateral ovary in two cases and in the contralateral ovary in another two. All cases had endometrioid ovarian carcinomas (EOC) of various types admixed with typical YST components. Immunohistochemically, EOC areas differed from YST in their positivity for OC 125, CA 19.9, and nuclear estrogen and progesterone receptors and in their negativity for AFP, which was conspicuously positive in the YST areas. The clinicopathological profile of ovarian endometrioid tumors with YST also differs from that of YST in that it occurs in the same age range as EOC, it shows coexistence of benign endometrioid lesions, and it has a poor response to chemotherapy. The histological pattern in transitional areas may be difficult to differentiate from “endometrioid-like” (enteroblastic) YST and clear cell tumors. Ovarian endometrioid tumors with YST component should be considered a variant of endometrial carcinoma. Its recognition is necessary in view of its unusually aggressive behavior and poor prognosis.

γδ T cells represent a minor T-cell subset that is mainly distributed in mucosal surfaces. Two distinct lymphomas derived from these cells have been recognized: hepatosplenic γδ T-cell lymphoma (HSTL) and primary cutaneous γδ T-cell lymphoma (PCGD-TCL). However, whether other anatomic sites may also be involved and whether they represent a spectrum of the same disease are not well studied. The lack of T-cell receptor (TCR)β expression has been used to infer a γδ origin when other methods are not available. We studied 35 T-cell tumors suspected to be γδ TCL using monoclonal antibodies reactive with TCR δ or γ in paraffin sections. We were able to confirm γδ chain expression in 22 of 35 cases. We identified 8 PCGD-TCLs, 6 HSTLs, and 8 γδ TCLs without hepatosplenic or cutaneous involvement involving mainly extranodal sites. Two such cases were classified as enteropathy-associated T-cell lymphoma, type II. The other γδ TCL presented in the intestine, lung, tongue, orbit, and lymph node. In addition, we observed 13 cases with mainly extranodal involvement that lacked any TCR expression ("TCR silent"). In all cases, a natural killer cell origin was excluded. In conclusion, the lack of TCRβ expression does not always predict γδ-T-cell derivation, as TCR silent cases may be found. The recognition of γδ TCL presenting in extranodal sites other than skin and liver/spleen expands the clinical spectrum of these tumors. However, non-HSTL γδ TCL do not seem to represent a single entity. The relationship of these tumors with either HSTL or PCGD-TCL requires further study.

Cancer cells all share the feature of being immersed in a complex environment with altered cell-cell/cell-extracellular element communication, physicochemical information, and tissue functions. The so-called tumour microenvironment (TME) is becoming recognised as a key factor in the genesis, progression and treatment of cancer lesions. Beyond genetic mutations, the existence of a malignant microenvironment forms the basis for a new perspective in cancer biology where connections at the system level are fundamental. From this standpoint, different aspects of tumour lesions such as morphology, aggressiveness, prognosis and treatment response can be considered under an integrated vision, giving rise to a new field of study and clinical management. Nowadays, somatic mutation theory is complemented with study of TME components such as the extracellular matrix, immune compartment, stromal cells, metabolism and biophysical forces. In this review we examine recent studies in this area and complement them with our own research data to propose a classification of stromal changes. Exploring these avenues and gaining insight into malignant phenotype remodelling, could reveal better ways to characterize this disease and its potential treatment.