Immune profiling of dedifferentiated liposarcoma and identification of novel antigens for targeted immunotherapy.

Dedifferentiated liposarcoma (DDLS) is an aggressive, recurring sarcoma with limited treatments. T-cell immunotherapies selectively target malignant cells, holding promise against DDLS. The development of successful immunotherapy for DDLS requires a thorough evaluation of the tumor immune microenvironment and the identification and characterization of targetable immunogenic tumor antigens. To assess the complexity of the human DDLS tumor immune microenvironment and to identify target antigens, we used the nCounter NanoString platform, analyzing gene expression profiles across 29 DDLS and 10 healthy adipose tissue samples. Hierarchical clustering of tumors based on expression of tumor inflammation signature genes revealed two distinct groups, consisting of 15 inflamed tumors and 14 non-inflamed tumors, demonstrating tumor heterogeneity within this sarcoma subtype. Among the identified antigens, PBK and TTK exhibited substantial upregulation in mRNA expression compared to healthy adipose tissue controls, further corroborated by positive protein expression by IHC. This data shows considerable inter-tumoral heterogeneity of inflammation, which should be taken into consideration when designing an immunotherapy for DDLS, and provides a novel targetable antigen in DDLS. The results of this study lay the groundwork for the development of a novel immunotherapy for this highly aggressive sarcoma.

A randomized, non-comparative phase 2 study of neoadjuvant immune-checkpoint blockade in retroperitoneal dedifferentiated liposarcoma and extremity/truncal undifferentiated pleomorphic sarcoma.

Based on the demonstrated clinical activity of immune-checkpoint blockade (ICB) in advanced dedifferentiated liposarcoma (DDLPS) and undifferentiated pleomorphic sarcoma (UPS), we conducted a randomized, non-comparative phase 2 trial ( NCT03307616 ) of neoadjuvant nivolumab or nivolumab/ipilimumab in patients with resectable retroperitoneal DDLPS (n = 17) and extremity/truncal UPS (+ concurrent nivolumab/radiation therapy; n = 10). The primary end point of pathologic response (percent hyalinization) was a median of 8.8% in DDLPS and 89% in UPS. Secondary end points were the changes in immune infiltrate, radiographic response, 12- and 24-month relapse-free survival and overall survival. Lower densities of regulatory T cells before treatment were associated with a major pathologic response (hyalinization > 30%). Tumor infiltration by B cells was increased following neoadjuvant treatment and was associated with overall survival in DDLPS. B cell infiltration was associated with higher densities of regulatory T cells before treatment, which was lost upon ICB treatment. Our data demonstrate that neoadjuvant ICB is associated with complex immune changes within the tumor microenvironment in DDLPS and UPS and that neoadjuvant ICB with concurrent radiotherapy has significant efficacy in UPS.

Evaluation of PD-L1 Expression in Undifferentiated Pleomorphic Sarcomas, Liposarcomas and Chondrosarcomas.

Immune checkpoint inhibitors (ICIs) such as PD1/PD-L1 blockers are an established treatment for many solid cancers. There are currently no approved ICIs for sarcomas, but satisfactory results have been seen in some patients with disseminated disease in certain histological types. Most studies on PD-L1 in sarcoma have used small specimens and there are no clear cutoff values for scoring. We investigated PD-L1 immunoreactivity in high-grade chondrosarcomas (CS), abdominal liposarcoma (LS) and undifferentiated pleomorphic sarcomas (UPS). In total, 230 tumors were stained with SP142 and SP263 assays and evaluated by two clinical pathologists. Immunoreactivity in tumor and immune cells was correlated with clinical outcome. Overall, ≥1% PD-L1 immunoreactivity in tumor cells was found in 11 CS, 26 LS and 59 UPS (SP142 assay) and in 10 CS, 26 LS and 77 UPS (SP263 assay). Most tumors exhibited ≤10% PD-L1 immunoreactivity, but a subset across all three subtypes had >50%. Kaplan-Meier survival analysis showed no significant difference in metastasis-free or overall survival in relation to PD-L1 immunoreactivity in tumor or immune cells for any subtype. As there is a lack of clinical data regarding PD-L1/PD-1 status and therapy response, it is not currently possible to establish clear cutoff values. Patients with high (>50%) PD-L1 immunoreactivity in tumor cells (TC) with the SP263 assay would be a logical group to investigate for potentially beneficial PD1/PD-L1-targeted treatment.

Characterising the immune microenvironment in liposarcoma, its impact on prognosis and the impact of radiotherapy.

BACKGROUND AND OBJECTIVES: Limited literature exists examining the immune microenvironment in liposarcoma, particularly with regard to the impact of radiotherapy. A major problem is the lack of scoring system for the tumour-infiltrating lymphocytes (TILs) in sarcoma. This study aims to describe the immune environment pre- and postradiotherapy and identify the optimal immune infiltrate scoring system for sarcoma. METHODS: Thirty-nine paired tissue samples (pre- and postradiotherapy) from patients with liposarcoma were scored by two pathologists for TILs using pre-existing systems (for breast cancer and melanoma) and compared for interobserver reliability. Immunohistochemical staining was performed for various immune markers. RESULTS: The TIL scoring system for breast cancer yielded perfect agreement (κ = 1.000). 21% of patients had increased TILs after radiotherapy, 87.5% of whom had dedifferentiated liposarcoma. Immune suppressor expression was increased frequently after radiotherapy (CD68 increased in 59.4%, PD-L1 increased in 25%). Immune effector expression (CD8) was unchanged in 84.4%. CONCLUSIONS: Breast cancer TIL scoring is reproducible in liposarcoma and has high interobserver reliability. Radiotherapy was observed to have a limited impact on immune effectors but seemed to have more impact in upregulating immune suppressors, suggesting radiotherapy may contribute to disease control through immunomodulatory effects. Dedifferentiated liposarcoma represents a uniquely responsive subtype.

Comprehensive Immuno-Molecular Profiles for Liposarcoma: Roles of Programmed Death Ligand 1, Microsatellite Instability, and PIK3CA.

BACKGROUND: Developing personalized strategies for cancer has shown good efficacies. METHODS: We assessed the molecular targets programmed death ligand 1 (PD-L1), microsatellite instability (MSI), and PIK3CA. Seventy-four patients with liposarcomas who underwent curative resection were assessed for PD-L1 expression in the tumor and tumor-infiltrating lymphocytes (TILs), mismatch repair proteins (MLH1, PMS2, MSH2, and MSH6) by immunohistochemistry, MSI using polymerase chain reaction, and PIK3CA mutation/amplification using pyrosequencing and fluorescence in situ hybridization. RESULTS: Seventeen (23%) cases were TIL+ (≥1 + expression) and associated with longer 5-year overall survival than those with TIL- tumors (84.4 vs. 60.8%, p = 0.007). Six (35.3%) PD-L1+ tumors were detected only in TIL+ cases, with none detected in tumor cells. Two well-differentiated liposarcomas showed MSI, one low and one high with concurrent loss of MLH1, MSH6, and PMS2. PIK3CA mutation was detected in 7 (9.5%) [exon 9 (n = 4) and exon 20 (n = 3)] and only 1 Q546K mutation was a PD-L1+ tumor. PIK3CA copy number gain was detected in 18 (24.4%) and was associated with TIL+ tumors (p = 0.045). CONCLUSIONS: Our comprehensive immuno-molecular panel suggests that liposarcoma should be categorized based on the molecular genomic subtype for precision medicine.

Anti-tumor effects of anti-PD-1 antibody, pembrolizumab, in humanized NSG PDX mice xenografted with dedifferentiated liposarcoma.

The efficacy of an immune checkpoint blockade has been demonstrated against various types of cancer, but its suitability has not been fully proven for therapies specifically targeting sarcoma. We conducted a pan-cancer tumor data analysis to identify key immune-related variables strongly associated with sarcoma prognosis, and we explored whether these expected factors are functionally correlated with anti-PD-1 therapy in humanized (Hu) NOD.Cg-PrkdcscidIl2rgtm1Wjl/SzJ (NSG) mice xenografted with dedifferentiated liposarcoma (DDLPS). We found that an abundance of hCD8+ T cells and hNK cells was functionally associated with anti-PD-1 effects in the Hu-NSG DDLPS mice. Phenotypically, these cells were shown to be hCD8+IFNγ+, hCD8+PD-1+, hCD8+Ki-67+, hCD56+IFNγ+, hCD56+PD-1+, and hCD56+Ki-67+ cells and were enriched in splenocytes and tumor-infiltrating lymphocytes (TILs) of Hu-NSG DDLPS mice treated with anti-PD-1 antibody. Moreover, a considerable increase in activated hCD56+NKp46+NKG2D+ NK cells was also detected. Our findings suggest that hCD8+ T and hNK subsets play a pivotal role in anti-DDLPS tumor effects of anti-PD-1 therapy. The results provide clinical reference for advanced anti-PD-1 therapy targeting sarcoma tumors including DDLPS.

Comprehensive immune characterization and T-cell receptor repertoire heterogeneity of retroperitoneal liposarcoma.

Retroperitoneal liposarcoma (RLPS) is one of the most common subtypes of retroperitoneal soft tissue sarcomas and lacks effective treatment. This study aimed to provide a thorough profile of immune characteristics of RLPS. This study included 56 RLPS patients. Multisite tumor tissues were collected from 16 patients. Immunohistochemistry was carried out to identify CD4+ , CD8+ , FoxP3+ , CD20+ , or programmed cell death-1 (PD-1)+ tumor infiltrating lymphocytes (TILs) and  Programmed cell death ligand-1 (PD-L1) expression in tumor tissues. Ultradeep sequencing of T-cell receptor (TCR) ß-chain gene was carried out in 42 tumor samples as well as peripheral blood samples collected from 6 patients. In RLPS, TILs were distributed in 3 patterns and T cells were more prevalent than B cells. Generally, the proportion of TILs decreased and PD-L1 expression increased with tumor progression. Patients with higher PD-1/PD-L1 expression tended to have poorer prognosis, whereas patients with tertiary lymphoid structure tended to have a favorable disease-free survival. Although T-cell clones in tumors were quite different from those in peripheral blood, TCR sequencing showed low TCR repertoire reads as well as polyclonal status within tumors, which indicated limited T cell response in the tumors. Both TILs distribution and TCR repertoires suggested spatial immune heterogeneity in RLPS. Our research described the immune landscape of RLPS, and suggested RLPS might be a kind of tumor with low T cell infiltration as well as great immune heterogeneity. Therefore, strategies that can facilitate lymphocytic infiltration and immune reactivity need to be developed in the future to improve the efficacy of immunotherapy.

An atypical pleomorphic lipomatous tumor arising on the cheek.

Pathologists and dermatopathologists commonly encounter tumors with adipocyte differentiation. Most are of minimal clinical significance. Those exhibiting atypical spindle cell morphology have been reported but the terminology for such neoplasms is unsettled. Tumors with both spindle cell and pleomorphic morphology are rare, with equally unsettled descriptive nomenclature. Recently, a series of such tumors, termed atypical pleomorphic lipomatous tumor, has been published. They resided in the deep soft tissue or subcutis. To date, such a tumor has not been reported with dermal involvement.

Immunoreactivity of Canine Liposarcoma to Muscle and Brown Adipose Antigens.

Liposarcoma, rhabdomyosarcoma, and hibernoma share some overlapping histologic and immunohistochemical features. Although immunohistochemistry (IHC) is commonly used in the diagnosis of these neoplasms, expression of muscle markers has been reported in human liposarcoma and canine hibernoma in addition to rhabdomyosarcoma. Thus, these neoplasms are a diagnostic challenge but important to distinguish because of differences in prognosis and treatment. Rhabdomyosarcoma and liposarcoma are both malignant, but rhabdomyosarcoma has a higher potential for metastasis. In contrast, hibernomas are benign with low risk of recurrence. This study investigated expression of the muscle markers desmin, myogenin, and α-smooth muscle actin (α-SMA) and the brown fat marker uncoupling protein 1 (UCP1) in 25 cases of canine liposarcoma using IHC. Oil red O histochemistry was performed to confirm the presence of lipid and the diagnosis of liposarcoma in cases that were not well-differentiated. The 25 cases included 15 well-differentiated, 5 pleomorphic, 3 myxoid, and 2 dedifferentiated subtypes of liposarcoma. By IHC, 23 of 25 expressed UCP1, 7 of 25 expressed α-SMA, 7 of 25 expressed desmin, and 3 of 25 expressed myogenin with no clear relationship of antigen expression and tumor subtype. These findings clarify the immunohistochemical profile of canine liposarcoma and suggest overlap in the expression of several muscle antigens and UCP1 between liposarcoma, hibernoma, and rhabdomyosarcoma.

Dendritic fibromyxolipoma: a variant of spindle cell lipoma with extensive myxoid change, with cytogenetic evidence.

Dendritic fibromyxolipoma (DFML), a rare, recently described distinct benign soft tissue tumor, has many clinicopathological features reminiscent of spindle cell lipoma and solitary fibrous tumor with myxoid change. It is distinguished histologically from both entities by the presence of spindle and stellate cells with dendritic cytoplasmic prolongations, prominent myxoid stroma with abundant keloidal collagen and occasional small plexiform vascular proliferation. We describe a case of histologically confirmed DFML of the left shoulder in a 67-year-old male, in which subsequent cytogenetic analysis revealed deletion involving 13q14.3 region in all the tumor cells, typically detected in spindle cell lipoma. In the presence of many clinicopathological similarities between DFML and spindle cell lipoma including chromosomal abnormalities, we postulate that DFML is merely a rare variant of spindle cell lipoma with extensive myxoid degeneration, and may not be considered as a separate entity. The possible differential diagnosis and their distinguishing features are briefly discussed.

Heterogeneity and immunophenotypic plasticity of malignant cells in human liposarcomas.

Liposarcomas are tumors arising in white adipose tissue (WAT) with avidity for local recurrence. Aggressive dedifferentiated liposarcomas (DDLS) may arise from well-differentiated subtypes (WDLS) upon disease progression, however, this key issue is unresolved due in large part to knowledge gaps about liposarcoma cellular composition. Here, we wished to improve insights into liposarcoma cellular hierarchy. Tumor section analysis indicated that the populations, distinguishable based on the expression of CD34 (a marker of adipocyte progenitors) and CD36 (a marker of adipocyte differentiation), occupy distinct intra-tumoral locations in both WDLS and DDLS. Taking advantage of these markers, we separated cells from a panel of fresh human surgical specimens by fluorescence-activated cell sorting (FACS). Based on chromosome analysis and the culture phenotypes of the composing populations, we demonstrate that malignant cells comprise four mesenchymal populations distinguished by the expression of CD34 and CD36, while vascular (CD31+) and hematopoietic (CD45+) components are non-neoplastic. Finally, we show that mouse xenografts are derivable from both CD36-negative and CD36-positive DDLS cells, and that each population recreates the heterogeneity of CD36 expression in vivo. Combined, our results show that malignant cells in WDLS and DDLS can be classified according to distinct stages of adipogenesis and indicate immunophenotypic plasticity of malignant liposarcoma cells.

Presence of adenovirus species C in infiltrating lymphocytes of human sarcoma.

Human adenoviruses are known to persist in T-lymphocytes of tonsils, adenoids and intestinal tract. The oncogenic potential of different adenovirus types has been widely studied in rodents, in which adenovirus inoculation can induce multiple tumors such as undifferentiated sarcomas, adenocarcinomas and neuroectodermal tumors. However, the oncogenic potential of this virus has never been proven in human subjects. Using a highly sensitive broad-spectrum qRT-PCR, we have screened a set of different human sarcomas including leiomyosarcoma, liposarcoma and gastro intestinal stroma tumors. Primers binding the viral oncogene E1A and the capsid-coding gene Hexon were used to detect the presence of adenovirus DNA in tumor samples. We found that 18% of the tested leiomyosarcomas and 35% of the liposarcomas were positive for the presence of adenovirus DNA, being species C types the most frequently detected adenoviruses. However, only in one sample of the gastro intestinal stroma tumors the virus DNA could be detected. The occurrence of adenovirus in the tumor sections was confirmed by subsequent fluorescence in-situ-hybridization analysis and co-staining with the transcription factor Bcl11b gives evidence for the presence of the virus in infiltrating T-lymphocytes within the tumors. Together these data underline, for the first time, the persistence of adenovirus in T-lymphocytes infiltrated in muscular and fatty tissue tumor samples. If an impaired immune system leads to the viral persistence and reactivation of the virus is involved in additional diseases needs further investigation.

Effect of thermal stress on MICA/B induction in a human liposarcoma cell line.

BACKGROUND: A possible mechanism by which hyperthermia enhances tumor immunogenicity is the induction of NKG2D ligands on tumor cells. Although the expression of MHC class I chain-related protein A and B (MICA/B) has previously been reported in different carcinomas, there is no information about MICA/B expression in liposarcomas. OBJECTIVE: To investigate MICA/B induction in a human liposarcoma cell line (SW-872) after thermotherapy. METHODS: SW-872 and HeLa cell lines were subjected to thermal stress for 1 h at 42, 44 and 46C, and after 2, 4 and 6 h of incubation at 37C, MICA/B expression was assessed at the mRNA and protein levels. RESULTS: Despite high levels of MICA/B transcripts in SW-872 cells at baseline, the expression of these genes decreased significantly at both the mRNA and protein levels after almost all thermal treatments. CONCLUSION: Our data conclude that thermotherapy under 42-46 C had no effect on MICA/B induction on SW-872 liposarcoma cell line but the effects of fever-range temperatures remain to be tested on this cell line.

Parosteal osteoliposarcoma: a new bone tumor (from imaging to immunophenotype).

INTRODUCTION: Parosteal osteosarcomas and well-differentiated liposarcomas (WDLPS) of soft tissue share several features: they are slowly progressive, locally aggressive tumors, tend to recur locally, and rarely or never metastasizes if not dedifferentiated. Their treatment is wide surgical resection. Microscopically, both are well differentiated tumors, very like their normal tissue counterpart. They share simple karyotypes with supernumerary ring chromosomes or giant marker chromosomes containing amplified 12q sequences including MDM2 and CDK4 genes, with subsequent overexpression of MDM2 and CDK4 proteins. We present the case of a parosteal osteoliposarcoma made of closely intermingled components of a low-grade osteosarcoma and a WDLPS. CASE: In a 34 year-old woman with a slowly growing mass of the arm, imaging revealed a large well-defined heterogeneous parosteal mass of the upper humerus, with two main components: bone at the base and fat at the periphery. Microscopically, these two components were consistent respectively with low grade osteosarcoma and WDLPS. Cells of the two components were labeled with anti-CDK4 antibody. No labeling with anti-MDM2 antibody and no signal detected with MDM2 FISH analysis were likely due overdecalcification. No frozen tumor tissue was available for FISH analysis nor array-CGH. DISCUSSION: Differential diagnoses of this new entity would be a well-differentiated liposarcoma with a low-grade osteosarcomatous component that originates from the soft tissues, ruled out on imaging, and an ossifying parosteal lipoma, ruled out on immunohistochemistry. CONCLUSION: This is the first description of a low-grade parosteal sarcoma with two components that morphologically and immunophenotypically demonstrate characteristics of a parosteal osteosarcoma and of a well-differentiated liposarcoma.

[Current idea of the morphogenetic features of liposarcomas].

The paper considers the current aspects of the morphology, immunohistochemistry, molecular biology of liposarcomas. Particular attention is given to the embryogenesis of liposarcomas and to problems in the immunohistochemical diagnosis of these neoplasms. There is evidence that TLS-CHOP is involved in the PNA processing of liposarcoma cells. Prospects for the diagnosis and treatment of malignant mesenchymal tumors are discussed.

Conversion to rapamycin immunosuppression for malignancy after kidney transplantation.

INTRODUCTION: Malignancies are a well-known complication of immunosuppressive therapy among renal transplant recipients, representing an important cause of long-term morbidity and mortality. Rapamycin has been shown to limit the proliferation of a number of malignant cell lines in vivo and in vitro. METHODS: Fifteen patients developed the following malignancies at a mean of 90.3 months (range = 10-252) after kidney transplantation: metastatic gastric cancer (n = 1), metastatic colon cancer (n = 1), bilateral nephrourothelioma (n = 1), skin cancer (n = 2), Kaposi's sarcoma (n = 2), posttransplant lymphoproliferative disorder (PTLD; n = 4), renal cell carcinoma T1 (n = 1), MALT lymphoma (n = 1), intramucous colon carcinoma (n = 1), liposarcoma of the spermatic cord (n = 1). After the diagnosis of malignancy, the patients were switched from calcineurin inhibitor-based immunosuppression to rapamycin (monotherapy, n = 3), or associated with steroids (n = 6) or with mycophenolate mofetil (n = 6). RESULTS: Both patients with metastatic cancer underwent chemotherapy but succumbed after 6 and 13 months. Two patients with PTLD who underwent chemotherapy died after 12 and 36 months. At a mean follow-up of 32.7 months (range = 7-56), the remaining 11 patients are cancer-free. Two patients lost their grafts after 24 and 36 months after the switch due to chronic rejection. Renal graft function remained stable in all other patients from diagnosis throughout follow-up. CONCLUSION: Our observations suggested that rapamycin-based immunosuppression offers the possibility for regression of nonmetastatic tumors. Nevertheless, it is difficult to assess whether tumor regression was due to rapamycin treatment or to the reduced immunosuppression.

Immunoreactivity of p53, mdm2, and p21WAF1 in dedifferentiated liposarcoma: special emphasis on the distinct immunophenotype of the well-differentiated component.

Alteration of the p53/mdm2 pathway has been reported in the well-differentiated liposarcoma (WDLS)/dedifferentiated liposarcoma (DDLS) group. We investigated the immunoreactivity of p53, mdm2, and p21WAF1, along with the MIB-1-labeling index (MIB-1-LI) in 21 WDLS and 21 DDLS cases, to clarify the association of these markers with the morphologic changes and the biological factors responsible for the aggressiveness of DDLS. Within DDLS, p53 and p21WAF1 expression and mdm2 overexpression were significantly more prevalent in the dedifferentiated (DD) components than in the well-differentiated (WD) components. The mdm2 overexpression and p21WAF1 expression was significantly associated with sclerosing liposarcomas in both WDLS and the WD components of DDLS. There was no significant difference in the immunoreactivity of p53, mdm2, or p21WAF1 or MIB-1-LI between WDLS and the WD components of DDLS. An association was found between p53 expression and mdm2 overexpression in the WD group (comprising WDLS and WD components of DDLS) and in the DD group, significantly so in the WD group. Notably, this correlation was found in the subtype of sclerosing liposarcoma but not in that of lipoma-like liposarcoma. Within DDLS, the clinical outcome of the nonaccessible soft tissue (non-AST: comprising retroperitoneum and mediastinum) group was significantly worse than that of the accessible soft tissue (AST: comprising extremities, buttocks, axilla, and scrotum) group; however, the immunophenotypes of p53, mdm2, and p21WAF1 and the MIB-1-LI showed no correlation with survival in the AST group alone, in the non-AST group alone, or in the 2 together. This study suggests that the immunoreactivity of p53, mdm2, and p21WAF1 is associated with the morphologic changes, but not with the biological factors responsible for the aggressiveness of DDLS.

Immunoreactivity for the human hematopoietic progenitor cell antigen (CD34) in lipomatous tumors.

The human hematopoietic progenitor cell antigen (CD34) recently was shown to react with a variety of nonhematopoietic tissues and their tumors, including vascular endothelium, dendritic interstitial fibroblastic cells, and endoneurial cells as well as with the neoplastic cells in a variety of mesenchymal neoplasms of unknown etiology, such as Kaposi's sarcoma, dermatofibrosarcoma protuberans, epithelioid sarcoma, gastrointestinal stromal tumors, and solitary fibrous tumors. Additionally, it has been claimed that normal adipocytes may also react with this antibody. We studied a series of 90 lipomatous lesions to examine the pattern of immunoreactivity of the CD34 antigen in adipose tissue neoplasms. The study included 14 lipomas, 19 angiolipomas, 4 atypical lipomas, 18 spindle cell lipomas, 3 renal angiomyolipomas, 1 intramuscular lipoma, and 31 liposarcomas. Immunostains identified a network of CD34+ spindle cells admixed with the adipose tissue elements in all cases of lipoma, angiolipoma, angiomyolipoma, intramuscular lipoma, and well-differentiated lipoma-like liposarcoma. Additionally, the spindle cell component in all cases of spindle cell lipoma were strongly positive for this antigen. Atypical, stellate spindle cells and multinucleated "floret" cells in all cases of atypical lipoma as well as in six of 12 cases of well-differentiated lipoma-like liposarcoma of deep soft tissue were also positive for CD34. Scattered spindle cells in all cases of myxoid liposarcoma and in one case of round cell liposarcoma, as well as the sarcomatous component in one case of "dedifferentiated" liposarcoma, were strongly positive for this antigen. The round cells in myxoid liposarcoma and round cell liposarcoma, the signet-ring and multivacuolated lipoblasts in well-differentiated liposarcoma, and the pleomorphic atypical cells in pleomorphic liposarcoma were uniformly negative. The results of this study appear to indicate that lipomatous tumors may harbor a population of CD34+ interstitial dendritic spindle cells. Overgrowth or clonal expansion of this dendritic cell subpopulation may account for the development of spindle cell lipomas and for the spindle cell component in some cases of "dedifferentiated" liposarcoma.

Expression of HLA-A,B,C, beta 2-microglobulin (beta 2m), HLA-DR, -DP, -DQ and of HLA-D-associated invariant chain (Ii) in soft-tissue tumors.

Non-neoplastic mesenchymal cells, along with 33 benign and 87 malignant soft-tissue tumors (STT) were examined for expression of HLA-A,B,C, beta 2-microglobulin (beta 2m), HLA-DR, -DP, and -DQ molecules and the HLA-D associated invariant chain (Ii). Serial frozen sections were immunostained using monoclonal antibodies (MAbs) to monomorphic framework determinants of HLA sublocus products, beta 2m and Ii, and to CD53, a recently defined broadly distributed pan-leucocyte molecule. Compared with the normal state, an induction/neo-expression of HLA-A,B,C/beta 2m was found in a considerable number of tumors of muscle, peripheral nerve, cartilage-forming, adipose, and vascular tissues. Conversely, some tumors of fibrous origin and of autonomic ganglia showed an abnormal abrogation/loss of HLA-A,B,C/beta 2m with respect to their cells of origin. Small, round tumor cells present in various types of STT exhibited a heterogenous pattern of expression of these molecules with a preponderance of HLA-A, B,C/beta 2m-negativity. HLA-D/Ii determinants were rarely detectable in STT. Besides their expression in some fibrohistiocytic tumors, they were only occasionally found in tumors of smooth-muscle, peripheral-nerve and vascular origin as well as in one clear-cell sarcoma. In all tumors but one, there was no microtopographic association between HLA-D/Ii-positive tumor cells and inflammatory cells. CD53 allowed discrimination between dendritic interstitial cells (DIC) and neoplastic cells and additionally revealed that, in contrast to other solid tumors, STT are generally characterized by an extreme scarcity of lymphohistiocytic infiltrates. Our data indicate that, aside from very rare exceptions, aberrant induction or abrogation of MHC molecules in STT occurs in the absence of lymphohistiocytic stromal infiltrates, suggesting that these alterations might not be a consequence of local cytokine effects.