To compare the sensitivity and suitability of detection of active cytomegalovirus (CMV) infection by using monoclonal antibodies against CMV antigen (antigen test to detect antigenemia) and the polymerase chain reaction (PCR; to detect viral DNA) in granulocytes, 19 heart and 2 lung transplant recipients were closely monitored by these tests for at least 3 months after transplantation. All patients were CMV seropositive or had a seropositive donor. In total, 201 samples were tested; 46 were positive by both tests, 9 samples showed only antigenemia, 54 samples were positive by PCR only, and 102 samples were negative by both tests. PCR was positive earlier after transplantation in eight patients, whereas antigenemia was positive earlier after transplantation in one patient. In another four patients, both tests were positive at the same time. PCR was, on average, positive for a longer period of time. Discordant results showing a positive antigen test and a negative PCR were partly due to sampling error; some were positive by PCR on retesting. Samples which were negative by the antigen test and positive by PCR were taken at the beginning or at the end of an active CMV infection. In two patients, no active CMV infection was detected by the antigen test, cultures of urine and saliva, or serology, although PCR was positive for a long period of time in the two patients.

SUMMARYIn order to study the importance ofthe immune status ofthe donor in the development of immunity after allogeneic bone marrow transplantation (BMT), we monitored 23 cytomegalovirus (CMV) antibody-positive BMT recipients for humoral and cellular immunity to CMV, of whom 12 had a CMV antibody-positive and 11 a CMV antibody-negative marrow donor. Lymphocyte proliferation to CMV recovered significantly earlier after BMT in recipients of marrow from a CMV+ donor (10-4 weeks after BMT) compared with the recipients of marrow from CMV" donors (16-7 weeks after BMT, /'<0-05). This seemed to be specific, as lymphocyte proliferation to phytohaemagglutinin and Candida were not different between the two groups. IgM responses after active infection were seen in both groups, but initial IgG rises without IgM were seen only in recipients of marrow from CMV+ donors (/'<0-05). Lymphocyte proliferative or humoral immune responses to CMV were not detected in any ofthe patients in a control group consisting of nine CMV" recipients. These results indicate that T cell memory to CMV is transferred with donor marrow from CMV+ donors, leading in most patients to direct IgG anti-CMV responses and to quicker recovery of cellular immunity to CMV.

SUMMARYIn this study, several factors influencing the occurrence of active CMV infection after organ transplantation (Tx) are analysed. For this purpose, 105 heart, kidney and lung transplant recipients who were CMV-positive or had a CMV-positive donor, were closely monitored for active CMV infection by antigenaemia, cultures, CMV serology and lymphocyte proliferation (LP) to CMV. Univariate and multivariate regression analysis were performed. As pretransplant risk factors the HLA-type and numbers of HLA mismatches between recipients and their donors, and the CMV serology of the recipient and donor were analysed. A new finding was that recipients of donors positive for HLA-B7 were especially at risk for developing active CMV infection (P= 0 03) and CMV disease (P= 0-03). This was not due to increased rejection treatment in these patients. Post-transplant risk factors for development ofactive CMV infection were absence ofdetectable cellular immunity to CMV (lymphocyte proliferation) after Tx (P< 001) and rejection treatment with OKT3 or ATG (P = 0-05). High levels of IgG anti-CMV did not prevent occurrence ofactive CMV infection or CMV disease in the CMV+ recipients.

A randomized study of prophylaxis with hyperimmune globulin (HIg) was performed in 28 cytomegalovirus (CMV)-seronegative heart and kidney recipients with CMV-seropositive donors who were extensively monitored for active CMV infection and CMV disease. Detection of CMV antigen in peripheral blood granulocytes (antigenemia) was the first sign of primary CMV infection, generally occurring several weeks before IgM or IgG anti-CMV antibodies were detected and before positive cultures appeared. A correlation was found between rejection treatment with OKT3 or ATG, severity of CMV disease, and graft loss. Rejection treatment had no influence on incidence of CMV transmission. Primary CMV infection occurred most often in older patients with older donors. No beneficial effects were seen with HIg prophylaxis, which was administered from week 1 until week 7 after transplantation. Incidence of primary CMV infection was equal in both groups (50%) and no influence on the severity of primary CMV infection was seen. The only effect that was seen was on the time from transplantation to detection of active CMV infection, which was prolonged by HIg prophylaxis.

SUMMARYThe activation of primary human T cells largely depends on the expression of both major histocompatibility complex (MHC) class 11 and B7 molecules on antigen-presenting cells (APC), whereas APC expressing HLA class I1 but not B7 antigens are expected to induce anergy. According to this concept, interferon-y (IFN-y)-activated keratinocytes (KC) expressing HLA class I1 but not B7 costimulatory antigens should be able to induce anergy. However, in terms of anergy versus'activation contradicting data have been published on the outcome of interaction between T cells and human KC. In addition, it has been shown that human KC can express a B7-like molecule with unknown function, whereas MHC expression may be functionally impaired. To evaluate this item we transfected the human A431 KC cell line with B7-1 coding sequences and upregulated HLA-DR by treatment with IFN-y, yielding A43 IDRsB7-' cells. Irradiated A43 lDRSB7-l cells were found to be capable of inducing vigorous proliferative primary T-cell responses in resting allogeneic T cells, whereas A43 lDR cells could induce proliferation only when interleukin-2 (IL-2) was added. These data indicate that KC can present alloantigens, and that lack of costimulatory molecules on KC is the main reason why these cells cannot induce primary T-cell responses. Surprisingly, however, no evidence could be obtained of stable anergy induction by A431DR cells, as T cells contacted with A431DR cells and then transferred to A43IDRsB7-' cells clearly demonstrated alloresponsiveness. T-cell non-responsiveness was maintained only when T cells remained in contact with A431DR cells. These data indicate that, despite expression of HLA class I1 in the absence of B7 costimulatory molecules, human KC cannot induce stable anergy but rather induce short-term anergy in primary resting T cells.