Immunological Risk Stratification of Bladder Cancer Based on Peripheral Blood Natural Killer Cell Biomarkers.

BACKGROUND: Bladder cancer (BC) is highly immunogenic. Bacillus Calmette-Guérin (BCG) immunotherapy offers the best results in non-muscle-invasive BC (NMIBC). Natural killer cells (NKcs) play decisive roles in BCG-mediated immune response and in general cancer immune-surveillance. OBJECTIVE: To analyze killer-cell immunoglobulin-like receptors (KIRs), their human leukocyte antigen class-I (HLA-I) ligands, and the expression of DNAX Accessory Molecule-1 (DNAM-1/CD226) on peripheral blood (PB) NKcs, to identify useful predictive biomarkers in BC. DESIGN, SETTING, AND PARTICIPANTS: KIR/HLA-ligand genotypes were compared between 132 BC, 201 other solid cancers, 164 plasma cell disorders, and 615 healthy Caucasoid controls. CD226 expression was evaluated by flow cytometry. OUTCOME MEASUREMENTS AND STATISTICAL ANALYSIS: KIR/HLA-I interactions and CD226 expression on NKcs (CD226high or CD226low) were compared across study groups, cancer stages, treatments, and progression-free and overall survival of patients, using chi-square, analysis of variance/post hoc, Kaplan-Meier/log-rank, and regression analyses. RESULTS AND LIMITATIONS: Three immunological risk groups were identified: low risk (KIR2DL1-L2+L3-/C1C1- and KIR2DL1+L2+L3+/C1C1+), intermediate risk (rest), and high risk (KIR2DL5+/HLA-C*16+ and KIR2DL1+L2+L3-), which displayed different 10-yr progression-free rates (83.3%, 48.6%, and 0%, respectively; p<0.001) and survival rates (83.3%, 54.3%, and 6.2%, respectively; p<0.001) for muscle-invasive T2/T4, and 10-yr progression-free rates (100%, 81.6%, and 50%, respectively; p<0.05) for NMIBC-T1 treated with BCG. Immunological risk stratification had an independent prognostic value to just histological staging for survival (hazard ratio=2.93, p<0.00001, Harrell C-statistic=0.779). CD226 expression on PB NKcs improved immunological stratification in intermediate-risk T1-T4 BC patients, with survival rates of 94.1% and 66.7% for CD226high and CD226low (p<0.05), respectively. CONCLUSIONS: Immunological risk stratification will complement BC histopathology to improve risk stratification and guide the selection of personalized treatments. Understanding of the molecular mechanisms of NKc tumor immune surveillance will enable the development of future NKc-based therapies. PATIENT SUMMARY: This work describes a peripheral blood test that aids in our understanding of the immune defense mechanisms against bladder cancer, is useful for classifying patient risk, and will guide personalized treatments.

Activating KIRs on Educated NK Cells Support Downregulation of CD226 and Inefficient Tumor Immunosurveillance.

Therapies using NK cells (NKc) expanded/activated ex vivo or stimulated in vivo with new immunostimulatory agents offer alternative opportunities for patients with recurrent/refractory tumors, but relevant biomarkers to guide the selection of patients are required for optimum results. Overall survival of 249 solid cancer patients was evaluated in relation to the genetics and/or the expression on peripheral blood NKcs of inhibitory and activating killer-cell immunoglobulin-like receptors (iKIR and aKIR, respectively), HLA class I ligands, CD226 (also known as DNAM-1), and NKG2A. Compared with patients with higher expression, patients with low expression of CD226 on total NKcs showed shorter mean overall survival (60.7 vs. 98.0 months, P < 0.001), which was further reduced in presence of telomeric aKIRs (KIR2DS1-DS5 and/or KIR3DS1, 31.6 vs. 96.8 months, P < 0.001). KIR2DL2/S2+, KIR3DL1+, KIR2DL1+, and KIR2DL3+ NKc subsets in the presence of their cognate ligands primarily contributed to shortening patients' overall survival by increasing the sensitivity to CD226 downmodulation in aKIR-rich telomeric genotypes. In patients with high tumor burden who died during the follow-up period, aKIR-rich telomeric genotypes were associated with: (i) specific downmodulation of CD226 on educated NKcs but not on CD8+ T cells or uneducated NKcs, (ii) lower expression of CD226 and higher expression of NKG2A on aKIR+ NKcs, and (iii) lower numbers of total CD56dim NKcs. The reduced expression of CD226 on NKcs with aKIR-rich genotypes may be a biomarker indicative of NKc hyporesponsiveness in patients that could benefit from new NKc immune-stimulatory therapies.

KIR gene mismatching and KIR/C ligands in liver transplantation: consequences for short-term liver allograft injury.

BACKGROUND: Killer immunoglobulin-like receptors (KIRs) bind human leukocyte antigen (HLA) class-I (HLA-I) ligands and regulate functions of natural killer cells and subsets of T cells. KIR/HLA-I interactions allow predicting natural killer cell alloreactivity in hematopoietic stem cell transplantation and in HLA-compatible kidney transplants, but its meaning in liver transplantation remains controversial. METHODS: KIR and HLA genotypes were studied in 402 liver transplants, using sequence-specific oligonucleotides and primer methods. Recipients and donor KIRs, HLA-C genotypes, KIR gene mismatches (MMs) between recipient-donor pairs, and KIR/HLA-ligand combinations were analyzed in overall transplantations, in the acute rejection (AR; n=110) and non-AR (n=292) groups. RESULTS: KIR gene MMs between recipients and donors, mainly in activating KIRs, and KIR2DL3 and KIR2DS1 of recipients in the presence of donor C2 ligands, significantly enhanced early AR rate (P<0.05), with KIR2DL3 and KIR2DS1 exhibiting a synergic effect in dependence of the donor C2 ligand number (χ2=7.662, P=0.022). KIR2DL3, KIR2DS1, and also KIR2DS4 significantly influenced short-term graft survival, with a benefit for transplantations combining KIR2DL3 recipients and donors having C1 ligands (log rank, P<0.019 at 1 year; hazards ratio [HR], 0.321; 95% confidence interval [CI], 0.107-0.962; P=0.042), whereas KIR2DS1 and KIR2DS4 recipients combined with donors lacking C1 ligands (C2/C2) exhibited a worse graft survival (log rank, P=0.035 at 6 months; HR, 7.713; 95% CI, 2.156-27.369; P=0.002 for KIR2DS1; and log rank, P=0.006 at 1 year; HR, 3.794; 95% CI, 1.267-11.365; P=0.017 for KIR2DS4). CONCLUSIONS: This study shows that KIR gene-gene MMs increase AR and that KIRs/C ligands associated to AR and KIR2DS4/C ligands also influence short-term graft survival.

KIR gene variability in cutaneous malignant melanoma: influence of KIR2D/HLA-C pairings on disease susceptibility and prognosis.

Natural killer and CD8(+) T cells are believed to be involved in the immune protection against melanoma. Their function may be regulated by a group of receptors defined as killer immunoglobulin-like receptors (KIRs) and their cognate HLA class I ligands. In this study, we analyzed the influence of KIR genes and KIR/HLA-I combinations on melanoma susceptibility and/or prognosis in a Spanish Caucasian population. For this purpose, KIR genotyping by PCR-SSP and HLA-C genotyping by reverse PCR-SSO were performed in 187 melanoma patients and 200 matched controls. We found a significantly low frequency of KIR2DL3 in nodular melanoma (NM) patients (P = 0.001) and in ulcerated melanoma patients (P < 0.0001). Similarly, the KIR2DL3/C1 combination was significantly decreased in melanoma patients (Pc = 0.008) and in patients with sentinel lymph node (SLN) melanoma metastasis (Pc = 0.002). Multivariate logistic regression models showed that KIR2DL3 behaves as a protective marker for NM and ulcerated melanoma (P = 0.02, odds ratio (OR) = 0.14 and P = 0.04, OR = 0.28, respectively), whereas the KIR2DL3/C1 pair acts as a protective marker for melanoma (P = 0.017, OR = 0.54), particularly superficial spreading melanoma (P = 0.02, OR = 0.52), and SLN metastasis (P = 0.0004, OR = 0.14). In contrast, the KIR2DL3(-)/C1C2 genotype seems to be correlated with NM and ulceration. We also report that the KIR2DL1(+)/S1(-)/C2C2 genotype is associated with susceptibility to melanoma and SLN metastasis. Altogether, the study of KIR2D genes and HLA-C ligands may help in assessing cutaneous melanoma risk and prognosis.

Low median fluorescence intensity could be a nonsafety concept of immunologic risk evaluation in patients with shared molecular eplets in kidney transplantation.

Human leukocyte antigen (HLA) antibodies are usually "epitope" and not "antigen" specific. This work presents an interesting case concerning Luminex median fluorescence intensity (MFI) levels in antibodies considered low risk (<1,000), but producing humoral rejection. These low-titer antibodies could play an important role in transplantation. A 42-year-old woman was retransplanted with a deceased donor with negative complement-dependent cytotoxicity cross-matching. Our patient was pretransplant (PrT) sensitized to HLA antigens (single antigens (SA) = 31%) for 1 previous transplant. Thus, the formerly detected sensitized antigens were A32, A30, A31, cross-reacting group 5C, and DQ3 with a MFI(max) ≈ 4,127. In the posttransplantation period (PTP), the patient exhibited important instability in renal function and we detected an increased SA percentage (61%) with MFI(max) = 15,029 (A*32) with other antigens (detected with a low PrT MFI [<1,000]) as anti-A*03 (MFI(max) = 13,301) and anti-A*11 (MFI(max) = 13,714) specificities. Anti-A*03 was a donor-specific antibody (DSA). Renal biopsy was compatible with humoral rejection. The patient was pulsed with methylprednisolone, plasmapheresis, and intravenous immunoglobulin without improvement. Thus, we added anti-CD20 and the initial clinical response was highly favorable. Biopsies resulted in suggestive rejection reversion. MFI A*03 DSA decreased to 6,908 and later to MFI(max) = 5,505. After a 6-month PTP, the patient is well with MFI(max) = 3,124. It was possible to define exactly the potential immunizing epitope eplets whose recognition determined the specific antibody production. A*32:01, A*30:01, A*31:01 (detected PrT), A*11:01, and A*03:01 (detected PTP) alleles have several shared eplets (62QE, 70AQS, and 76VGT), with 62QE being the only eplet present on all alleles. In conclusion, low MFI levels in antibodies considered low risk could be important in posttransplant humoral rejection, although the patient's renal function can be restored. Thus, specific shared eplets should always be investigated with respect to previous transplant mismatches.

HBV and HCV infections and acute rejection differentially modulate CD95 and CD28 expression on peripheral blood lymphocytes after liver transplantation.

Hepatitis B virus (HBV) and hepatitis C virus (HCV) generally reinfect liver graft early posttransplantation and lead to poorer graft and patient survivals. In the present study the influence of acute rejection (AR), HBV and HCV infections, and human leukocyte antigen (HLA) class-I compatibility on the expression of CD28 (in 237 liver recipients) and CD95 (in 114 liver recipients) on peripheral blood cells were evaluated by flow cytometry during the first month after transplantation. HBV/HCV infections induced strong CD95 upregulation on CD3+ lymphocytes. Maximal CD95 upmodulation was found in infected recipients showing partial HLA class-I compatibility. AR and virus reinfection could be distinguished because CD28 was upregulated on CD4+ lymphocytes only in recipients with AR, irrespective of their status regarding HBV/HCV infections. In conclusion, cytometric co-evaluation of CD95 and CD28 expression on peripheral blood lymphocytes could be useful to discriminate AR from cellular activation induced by viral reinfection of the liver graft.