Somatic mutations in the variable regions of a human IgG anti-double-stranded DNA autoantibody suggest a role for antigen in the induction of systemic lupus erythematosus.

The processes that govern the generation of pathogenic anti-DNA autoantibodies in human systemic lupus erythematosus (SLE) are largely unknown. Autoantibodies may arise as a consequence of polyclonal B cell activation and/or antigen-driven B cell activation and selection. The role of these processes in humoral autoimmunity may be studied by molecular genetic analysis of immunoglobulin (Ig) variable (V) regions of antibodies that are characteristic of SLE. We have analyzed the gene elements that encode a high affinity, IgG anti-double-stranded DNA autoantibody secreted by a monoclonal Epstein-Barr virus (EBV)-transformed cell line derived from a patient with active SLE. In addition, we have identified, cloned, and sequenced the germline counterparts of the VH and VL genes expressed in this autoantibody. The comparison of both sets of gene elements shows that the autoantibody VH and VL regions harbor numerous somatic mutations characteristic of an antigen-driven immune response. The light chain expressed in this autoantibody is a somatically mutated variant of the kv325 germline gene that is frequently associated with paraproteins having autoantibody activity and with Ig molecules produced by malignant B cells that express the CD5 antigen. Furthermore, the utilized DH segment has been repeatedly found in multireactive, low affinity IgM anti-DNA autoantibodies from SLE patients and healthy individuals. These results suggest that pathogenic IgG anti-DNA autoantibodies in human SLE may arise through antigen-driven selection of somatic mutations in the gene elements that frequently encode multireactive IgM autoantibodies.

Autoantibodies encoded by the most Jh-proximal human immunoglobulin heavy chain variable region gene.

Little is known about the utilization of human Ig heavy chain variable gene segments (VH segments) in different B-lineage cell populations or in antibodies of particular specificity and function. We now demonstrate that human antibodies with Ig VH regions encoded by the most JH-proximal human VH segment (VH6) have specificities resembling those of autoantibodies present in sera of patients with systemic lupus erythematosus (e.g., anti-DNA and anticardiolipin). These specificities appear to be encoded by the germline VH6 gene because the activity was found in multiple independent VH6 antibodies in which the light chain varied with respect to isotype and V kappa subgroup. Features of CDR3 length and somatic mutation patterns in several VH6 antibodies suggested that they were selected by the immune system.

T cells in cardiac allograft vasculopathy are skewed to memory Th-1 cells in the presence of a distinct Th-2 population.

Cardiac allograft vasculopathy (CAV) in heart transplantation (HTx) patients remains the major complication for long-term survival, due to concentric neointima hyperplasia induced by infiltrating mononuclear cells (MNC). Previously, we showed that activated memory T-helper-1 (Th-1) cells are the major component of infiltrating MNC in coronary arteries with CAV. In this study, a more detailed characterization of the MNC in human coronary arteries with CAV (n = 5) was performed and compared to coronary arteries without CAV (n = 5), by investigating MNC markers (CD1a, DRC-1, CD3, CD20, CD27, CD28, CD56, CD68, CD69, FOXP3 and HLA-DR), cytokines (IL-1A, 2, 4, 10, 12B, IFN-gamma, and TGF-beta1), and chemokine receptors (CCR3, CCR4, CCR5, CCR7, CCR8, CXCR3 and CX3CR1) by immunohistochemical double-labeling and quantitative PCR on mRNA isolated from laser microdissected layers of coronary arteries. T cells in the neointima and adventitia of CAV were skewed toward an activated memory Th-1 phenotype, but in the presence of a distinct Th-2 population. FOXP3 positive T cells were not detected and production of most cytokines was low or absent, except for IFN-gamma, and TGF-beta. This typical composition of T-helper cells and especially production of IFN-gamma and TGF-beta may play an important role in the proliferative CAV reaction.

Stem cell-derived cardiomyocytes after bone marrow and heart transplantation.

Cardiomyocytes are a stable cell population with only limited potential for renewal after injury. Tissue regeneration may be due to infiltration of stem cells, which differentiate into cardiomyocytes. We have analysed the influx of stem cells in the heart of patients who received either a gender-mismatched BMT (male donor to female recipient) or a gender-mismatched cardiac transplant (HTX; female donor to male recipient). The proportion of infiltrating cells was determined by Y-chromosome in situ hybridization combined with immunohistochemical cell characterization. In BM transplanted patients and in cardiac allotransplant recipients, cardiomyocytes of apparent BM origin were detected. The proportions were similar in both groups and amounted up to 1% of all cardiomyocytes. The number of stem cell-derived cardiomyocytes did not alter significantly in time, but were relatively high in cases where large numbers of BM-derived Y-chromosome-positive infiltrating inflammatory cells were present. The number of Y-chromosome-positive endothelial cells was small and present only in small blood vessels. The number of BM-derived cardiomyocytes in both BMT and HTX is not significantly different between the two types of transplantation and is at most 1%.

TNFalpha in patients with end-stage heart failure on medical therapy or supported by a left ventricular assist device.

BACKGROUND: In the heart elevated levels of TNFalpha can cause lethal heart failure, like Dilated Cardiomyopathy (DCM). The level of TNFalpha production is in part determined by promoter gene polymorphisms. We investigated whether the TNFalpha promoter gene polymorphism is in this way involved in the outcome of end-stage heart failure and predicts whether patients require left ventricular assist device (LVAD) support or can be kept on medical therapy (MT)while awaiting heart transplantation (HTx). As most patients in this study received a heart transplant, the role of the TNFalpha polymorphisms in transplant rejection was studied as well. METHODS AND RESULTS: In twenty nine patients with DCM, 35 patients with Ischemic Heart Disease (IHD; both on MT), 26 patients on LVAD support and 61 cardiac transplant donors TNFalpha plasma level was detected by EASIA. In both patients groups high levels of TNFalpha plasma levels was observed however, in patients supported by LVAD this increase was much higher compared to patients on MT. Furthermore, this increase seems to be associated with the TNF 1 allele ('G' at position -308) instead of the TNF2 allele (A at position -308). The promoter polymorphisms at positions -238, -244 and -308 were observed by polymerase chain reaction and sequencing. Polymorphism at positions -238, -244 and -308 did not show any relevant differences between the groups. However, at position -308, a trend of a higher incidence of the TNF2 allele (an "A" at position -308) in DCM patients compared to donors was shown. The distribution of the TNF1 and TNF2 alleles was not different in patients on medical therapy compared to the patients supported by a LVAD. No association was found between patients' TNFalpha promoter gene polymorphism and rejection. However, patients that received a donor heart with the TNF2 allele developed more rejection episodes, compared to patients that received a donor heart with the TNF1 allele. CONCLUSION: TNFalpha levels are high in patients with end-stage heart failure on MT, but even higher in patients on LVAD support. These high TNFalpha plasma levels however, are not correlated with the TNF2 allele but seems to be associated with the TNF1 allele. Furthermore, in HTx the donor TNFalpha gene seem to play a more important role in severity of acute rejection than that of the patient.

Implantation of cultured thymic fragments in patients with acquired immunodeficiency syndrome.

Cultured thymic fragments were implanted in one patient with acquired immunodeficiency syndrome (AIDS)-related complex (ARC) and in eight AIDS patients with opportunistic infections (OIs, four patients), Kaposi's sarcoma (KS, two patients), or both (two patients). Thereafter, objective clinical improvement was noted in one patient with OI, and a stable symptom-free condition was observed in the ARC patient and in two other patients with OIs. However, the ARC patient and two of the three patients with OIs developed infections three to six months after implantation. A fourth case of OI and the patients with KS showed progression of the disease. Peripheral blood investigations for counts of total leukocytes, lymphocytes, and T-lymphocyte subsets as well as for lymphocyte stimulation with mitogens showed no changes interpretable as an improvement of the cellular immune deficiency status. We conclude that cultured thymic fragments have no distinct in vivo effect on the course of AIDS, except for a temporary clinical improvement or a period of stable condition in some patients with OIs.

Influence of cytomegalovirus infection on the recovery of humoral immunity after autologous bone marrow transplantation.

Recovery of B-cell number and function was studied in 23 patients with hematological malignancies treated with high-dose chemoradiotherapy followed by autologous bone marrow transplantation (auto-BMT) in relation to the presence or absence of cytomegalovirus (CMV) infection. B cells recovered rapidly after auto-BMT and specific antibodies to herpes viruses remained nearly unchanged. Both were independent of the CMV status of the patients. However, the capacity of peripheral blood B cells to differentiate in vitro into cytoplasmic immunoglobulin (Ig)-positive cells (plasma cells) on pokeweed mitogen stimulation in the presence of normal T-cell help was significantly better in CMV-negative patients than in CMV-positive patients after auto-BMT, but was decreased in both groups. Serum Ig levels were, in contrast, higher in CMV-positive patients than in CMV-negative patients after auto-BMT.

Putative role of antireticulin antibody in antiacetylcholine-receptor-antibody-negative myasthenia gravis.

It has recently been demonstrated that pathogenic immunoglobulins circulate in the blood of patients with acetylcholine-receptor-antibody (A-AChR)-negative myasthenia gravis (MG). Evidence has been presented that in this form of MG the neuromuscular transmission is impaired by antibodies that bind to endplate determinants other than the AChR. We describe three patients with clinical manifestations of A-AChR-negative MG in whom antibody directed to reticulin (A-Ret) was detected. Antibody directed to reticulin is usually associated with celiac disease; however, none of the patients had symptoms or signs of celiac disease. To our knowledge, the association of A-Ret with A-AChR-negative MG has not been reported before. We postulate that A-Ret might help to differentiate between A-AChR-negative MG and congenital myasthenia. Further studies are needed to determine whether A-Ret plays a pathogenic role in A-AChR-negative MG or should instead be considered as an epiphenomenon.

Endomyocardial biopsies after heart transplantation. The presence of markers indicative of activation.

A series of 104 endomyocardial biopsies (EMB) from patients after heart transplantation was evaluated for the presence of immunological markers on graft component and infiltrating cells. This included markers for cells expressing alpha beta-T-cell receptors and gamma delta-T-cell receptors, and cytotoxic T cells with granules bearing the serine esterase Granzyme B; the presence of activation markers identified by CD25 (interleukin 2 receptor), CD30, CD69 (activation inducer molecule), CDw70; macrophages using antibody CD14 (WT14), and cells with Fc gamma-receptors type III (CD16). Almost all cells in T-cell infiltrates expressed the alpha beta-T cell receptor. Cells bearing the gamma delta-T cell receptor were scarcely found. The analysis with respect to the histopathologic diagnosis for rejection showed an absence of significance for T cell subsets, Granzyme B-positive cells, and activation markers except CD25. The numbers of macrophages labeled by CD14 and cells expressing Fc gamma RIII showed a significant relation to histopathology of rejection. Apart from leukocytes, also endothelium in EMB with rejection was labeled by the two anti-Fc gamma RIII antibodies used. In addition, in a small series of biopsies investigated, Fc gamma RI- and Fc gamma RII-positive cells were increased in EMB with rejection, and endothelium was labeled by Fc gamma RII antibodies. A cluster analysis on the basis of scores for CD25, CD14, and anti-Fc gamma RIII revealed three main clusters, one cluster comprising biopsies without abnormalities, one cluster containing EMB with the histopathology of rejection and high scores in immunophenotyping for lymphocytes and macrophages, and one cluster in between. The present data emphasize the importance of macrophage assessment in evaluating pathologic processes during rejection of heart allografts and diagnosing rejection.

Reduction of donor-specific cytotoxic T lymphocyte precursors in peripheral blood of allografted heart recipients.

Tolerance to allografted hearts in human recipients has been observed both in clinical situations and in in vitro experiments. To elucidate whether a quantitative change of alloreactive CTL is one of the mechanisms accounting for this graft tolerance, CTL precursor (CTLp) frequencies in the peripheral blood of 10 heart recipients were measured against spleen cells from donors and HLA nonidentical third-party persons. In this longitudinal follow-up study, we showed that the rejection reaction(s) in the grafted heart correlated with CTLp frequencies in samples taken before transplantation against the donor spleen cells, but not with the CTLp frequencies against the spleen cells from the third-party persons. The CTLp frequencies against the spleen cells from donors decreased 4-6 months after transplantation, and remained at a low level afterward. However, the CTLp frequencies against spleen cells from third-party persons in blood samples obtained 1 year after transplantation were not significantly different from those before transplantation. Therefore, we conclude that donor-reactive CTLs are important in rejecting allografted heart. The decrease in donor-specific CTLp after transplantation could explain the donor-specific tolerance. The decrease may be due to homing of the specific CTLp to the graft, or by clonal deletion of the donor-reactive CTL caused by chronic alloantigen stimulation in the presence of immunosuppressive therapies.

Enumeration of (auto)antibody producing cells in human using the "spot-ELISA".

An enzyme-linked immunosorbent assay (spot-ELISA) for individual immunoglobulin secreting cells, which became recently available, was applied to the enumeration of human B lymphocytes secreting specific antibodies of thyroglobulin. Polyclonally activated B cells from patients with auto-immune thyroid disease are incubated in thyroglobulin coated plates. After removal of the cells specific antibodies are visualized by means of an immunoenzyme technique employing agarose to localize converted substrate. Individual specific antibody secreting cells are counted as blue spots using an inverted microscope. Numbers and isotype of spots correlate well with the amount and isotype of secreted antibody as detected with a conventional ELISA. This easy to perform, complement-independent technique offers a useful alternative to conventional plaque forming cell assays.

Endothelial and smooth muscle cells in the heart allograft response: isolation procedure and immunocytochemical features.

Studies on mechanisms for allograft rejection are focused on recognition of major histocompatibility complex (MHC) antigens. In addition, there is evidence for non-MHC-mediated alloreactivity, possibly evoked by tissue-specific antigens. To measure cellular immune responses toward tissue-specific alloantigens, we isolated endothelial cells and smooth muscle cells from small pieces of human atrium at the time of transplantation. Endothelial cells were scraped off the endocardium after trypsin digestion and cultured in fibronectin-coated dishes. Smooth muscle cells were obtained by outgrowth of small pieces of atrium in a culture flask. Morphologic and immunologic characterization showed only minor differences between endothelial and smooth muscle cells cultured from atrium and cells cultured from umbilical vein (endothelial cells) and artery (smooth muscle cells). Furthermore, we studied the proliferative immune responses with endothelial and smooth muscle cells as stimulator cells, with and without induction of MHC class II antigens on these cells by addition of interferon-gamma to the culture. Peripheral blood mononuclear cells showed a proliferative response to donor human atrium endothelial cells, even without pre-incubation with interferon-gamma. Human atrium smooth muscle cells caused only a weak triggering of the mononuclear cells, irrespective of interferon-gamma pre-incubation. Immunofluorescence studies demonstrated HLA-DR expression on these endothelial and smooth muscle cells. These observations may indicate a role for non-MHC, probably tissue-specific, alloantigens expressed by endothelial cells in human cardiac allograft rejection.

Cytoimmunologic monitoring as an adjunct in monitoring rejection after heart transplantation: results of a 6-year follow-up in heart transplant recipients.

The cytoimmunologic monitoring assay has been proposed as a useful noninvasive technique in the diagnosis of rejection and infection after heart transplantation. In this study, we have analyzed the diagnostic usefulness of cytoimmunologic monitoring in 73 patients after heart transplantation. For individual patients, the follow-up varied between 2 and 78 months. Data were related to histopathologic characteristics of the endomyocardial biopsy. Significantly different cytoimmunologic monitoring results were not observed between groups according to endomyocardial biopsy histopathologic evaluation. The diagnostic usefulness of cytoimmunologic monitoring depended on the cutoff value applied. With higher cutoff values, the sensitivity decreased and the specificity and predictive value increased. For the previously reported cutoff value of 5%, the sensitivity was 0.29, the specificity was 0.73, and the predictive value was 0.66. Values of sensitivity, specificity, and predictive value were similar when only the first acute rejection was taken into account, or when only data on the first 4 weeks and the first 6 months after transplantation were considered. In calculating the diagnostic usefulness of the sensitivity, specificity, and predictive values were observed. We concluded that cytoimmunologic monitoring has a limited value for diagnosing acute rejection after heart transplantation.

Cytokine messenger RNA expression by donor-specific cytotoxic T-cell clones after allogeneic heart transplantation.

BACKGROUND: After heart transplantation, expression of various cytokines can be detected in endomyocardial biopsy specimens both in the presence and in the absence of rejection. In this study we have analyzed the contribution of donor-specific CD8+ cytotoxic T cells to intragraft cytokine expression. METHODS: T cells were propagated from endomyocardial biopsy specimens in medium containing interleukin-2 and interleukin-4. From the T-cell lines obtained, T-cell clones were generated by limiting dilution. A number of 15 CD8+ donor-specific cytotoxic T-cell clones were generated from a single T-cell line and were analyzed for their cytokine messenger RNA expression. The cytokine profile of the clones was studied on the mRNA level by reverse transcriptase polymerase chain reaction. RESULTS: Almost all clones expressed mRNA for interleukin-1 beta, interleukin-4, and interleukin-10, and 75% of the clones expressed mRNA for interleukin-1 alpha, interleukin-9, and tumor necrosis factor-beta. In about half of the clones expression of interleukin-2, interleukin-6, interleukin-8, and interferon-gamma was detected. Tumor necrosis factor-alpha was only detected in one of the clones. The cytokine profiles exhibited a considerable heterogeneity. The 15 clones had previously been analyzed for their T-cell receptor V beta-gene family expression and T-cell receptor V-D-J region sequence. Homology in the V-D-J regions of the clones indicated that the 15 clones were derived from a limited number of progenitor cells. Interestingly, clones derived from one progenitor expressed a different mRNA cytokine profile. CONCLUSIONS: This study shows that donor-specific cytotoxic T cells can contribute to the spectrum of locally produced cytokines. The cytokine expression of these cytotoxic T cells seems not to be limited to a distinct profile.

Cytotoxic T lymphocytes infiltrating the human cardiac allograft show a restriction in T-cell receptor V beta gene usage: a study on serial biopsy and blood specimens.

T lymphocytes were propagated in vitro from endomyocardial biopsy specimens that were obtained weekly from four patients during the first 2 to 3 months after heart transplantation. The culture was performed in the presence of recombinant interleukin-2 and interleukin-4, with or without mitogen, in which especially CD8+ donor-specific cytotoxic T cells expanded. These cells, presumably reflecting an in vivo activated population, could even be cultured from biopsy specimens without histopathologic signs of rejection. A preferential expression of T cell receptor V beta gene families was found in these T-cell lines. This finding is in contrast with the heterogenous expression in peripheral blood T cells of the same patient. The restriction in V beta gene family expression was substantiated in the evaluation of clones obtained from two cell lines. Among 17 donor-specific cytotoxic T-cell clones derived from one cell line, only four V beta gene families were expressed. All five clones from the other cell line used the V beta 8 family. Some clones expressing a distinct V beta gene family used the same V-D-J junction sequence, indicative of their origin from the same precursor. With the use of oligonucleotide probes complementary to clone-specific V-D-J junction sequences, four of five clones were detected not only in the parent T-cell line but also in T-cell lines derived from biopsy specimens with rejection reactions taken 1 week earlier and 2 weeks later and in blood cells taken before and 0.5, 3, and 6 months after transplantation; these clones were not detected in blood cells harvested 12 months after transplantation. This study showed a restricted usage of the V beta gene families by activated donor-specific cytotoxic T lymphocytes in the heart transplant.

Search for the antigen-specificity of homogeneous IgG components (H-IgG) after allogeneic bone marrow transplantation.

After allogeneic BMT, transient homogeneous Ig components (H-Ig) can be detected in the sera of most graft recipients. So far, data on the antigen-specificity and therefore the function of these H-Ig are not available. Such information may be important for our understanding of the underlying mechanisms that are responsible for these excessive clonal B cell expansions, and it may help to delineate the functional antibody repertoire after BMT. In the present study, sera of 98 paediatric BM graft recipients were investigated for the potential presence of H-Ig of IgG isotype (H-IgG) with specificity towards a panel of antigens, including vaccine and herpes virus antigens, auto-antigens and allo-antigens. The vast majority of H-IgG in sera of BM graft recipients were unreactive when tested for this panel of antigens. However, in four cases, antigen-specificity of H-IgG to tetanus toxoid could be demonstrated after vaccination with that antigen. An explanation for the negative findings may be either that a restricted antibody production had been elicited by other non-tested antigens, eg substances of colonizing and translocating bacteria or of food antigens, or that the H-IgG components may have anti-idiotype or anti-'self' specificity.

Monitoring rejection after heart transplantation: cytoimmunological monitoring on blood cells and quantitative birefringence measurements on endomyocardial biopsy specimens.

Cytoimmunological monitoring and quantitative birefringence measurements were used as potential aids in diagnosing acute rejection after heart transplantation instead of histopathological assessment of the endomyocardial biopsy specimen alone. Cytoimmunological monitoring was based on morphological inspection and quantitation of mononuclear cells, particularly activated lymphoid cells. Quantitative birefringence measurements comprise a variable for myocyte contractile function. Its read out is the ratio of the degree of birefringence before contraction to that after. Cytoimmunological monitoring indicated significantly higher concentrations of activated lymphocytes in moderate or severe acute rejection, and quantitative birefringence measurements indicated decreased myocyte function during severe and resolved or resolving rejection. Cytoimmunological monitoring and quantitative birefringence measurements were diagnostically most useful in terms of sensitivity, specificity, and predictive value, when only data gathered before the first episode of acute rejection were considered. For cytoimmunological monitoring, diagnostic relevance was optimal when the data were expressed as relative proportions of activated lymphocytes. The quantitative birefringence measurements correlated best with analysis of the endomyocardial biopsy specimen when a cut off value of 1.25 was used. When both methods for diagnosing acute rejection were analysed together, no improvement in sensitivity (value 0.44) was found, but the specificity increased to 0.98 and the predictive value to about 0.80. It is concluded that cytoimmunological monitoring is a useful, non-invasive additional method for diagnosing the first period of acute rejection after heart transplantation and that quantitative birefringence measurements give valuable information on the extent of myocyte damage.

Isozyme distribution of hexokinase, phosphofructokinase and pyruvate kinase in lymphocytes from patients with chronic lymphocytic leukemia.

The enzyme activities and isozyme distribution of the three glycolytic regulator enzymes hexokinase, phosphofructokinase and pyruvate kinase were studied in lymphocytes of patients with chronic lymphocytic leukemia. Isozyme distribution patterns were determined by kinetic measurements, electrophoresis and immunoprecipitation. The CLL lymphocytes were different from normal non-T lymphocytes with respect to hexokinase residual activity in the presence of glucose-1,6-P2, pyruvate kinase residual activity in the presence of alanine, and phosphofructokinase activity after stimulation by glucose-1,6-P2. No differences could be discerned in enzyme activities between the CLL and the normal T and non-T lymphocytes.

The prognostic value of the antiperinuclear factor, anti-citrullinated peptide antibodies and rheumatoid factor in early rheumatoid arthritis.

OBJECTIVE: To study the prognostic value of the antiperinuclear factor (APF), determined by an indirect immunofluorescence test (IIF) and a recently developed anti-citrullinated cyclic peptide (CCP) ELISA, in combination with rheumatoid factor (RF) status, in early RA (< 1 year). METHODS: A total of 249 participants in a randomized trial of treatment strategies were divided into 4 groups according to their APF (or CCP) and RF status at baseline. Differences in disability, joint involvement and radiological damage over a 3-year period were analysed. RESULTS: APF-IIF results differed from CCP-ELISA in 42 cases (17%); 38 of the 42 had a positive IIF and negative ELISA value. Disability after 3 years did not differ significantly between the RF and APF groups. APF- patients had significantly lower Thompson joint scores compared to APF+ patients (6 vs 24 for CCP-ELISA; 2 vs 24 for IIF). RF+APF+ patients exhibited more radiological damage compared to RF-APF- patients. RF+APF- and RF-APF+ patients had intermediate scores. Within the RF+ and RF- groups, APF+ was associated with more radiological damage and thus yielded prognostic information in addition to RF. In this respect, the results of ELISA and IIF were comparable. Thirty percent of the RF+APF+ patients had a radiological score higher than 45, compared to 13% of the RF+APF-, none of the RF-APF+, and 2% of RF-APF- patients (p < 0.001). In addition, more large joints were affected in APF+ than in APF- patients, while no difference was observed between RF+ and RF- patients. CONCLUSION: APF has prognostic value in addition to RF for joint involvement and radiological damage in early RA. The CCP-ELISA technique for APF assessment may facilitate its use in clinical practice. However, the prognostic value of the two tests lies in their ability to predict mild disease. Reliable identification at baseline of individual patients with progressive disease is still not possible.

Actinic reticuloid simulating Sézary syndrome. Report of two cases.

A report is given on two male patients who showed all the main characteristics of Sézary syndrome (SS). When phototested, however, they proved to be extremely photosensitive, which suggested a diagnosis of actinic reticuloid (AR). This was supported by the predominance in blood and skin specimens of lymphoid cells with a suppressor/cytotoxic phenotype, the absence of clonal cell proliferation and a benign clinical course. Differential diagnostic problems of SS and erythrodermic AR are discussed.