Randomized study of recombinant interleukin-2 after autologous bone marrow transplantation for acute leukemia in first complete remission.

Immunological control of acute leukemia may be achieved after allogeneic transplant. Despite promising preliminary results, the impact of immunotherapy with interleukin-2 (r-IL-2) on patients with acute leukemia (AL), in first complete remission (CR1) remains unclear. We conducted a prospective multicenter randomized trial to compare outcome in patients with AL in CR1, treated with autologous bone marrow transplantation (BMT) with or without postgraft r-IL-2. One hundred and thirty patients with AL in CR1 (myeloblastic (AML): N = 78; lymphoblastic (ALL): N = 52) were randomized at time of BMT to receive (N = 65) or not (N = 65) r-IL-2. r-IL-2 (RU 49637 from Roussel Uclaf) was started after hematological recovery, as a five cycle regimen (12 M IU/m2/day continuous infusion on day 1-5, 15-17, 29-31,43-45 and 57-59). The two groups were balanced for patient and transplant characteristics. Analysis was based on an intent to treat. Thirty-eight (59%) of the 65 patients randomized into the study group started r-IL-2 at a median of sixty-eight days (23-140) after transplant and received 77% (16-100) of the scheduled dosage. They received a median of 120 x 10(6) IU/m2 (25-156) over 10 (3-13) days during a total median period of 56 (3-78) days. With a median follow-up of 7 years (5.4-8.1 years), 79 patients relapsed (study group: 43 (66%); control group: 36 (55%): p = NS). Survival and leukemia-free survival estimates were 33% (23-45) versus 43% (22-52) and 29% (19-41) versus 36% (24-51) respectively for study and control groups (all p = NS). These results show that leukemic control after autologous BMT is not increased by r-IL-2 therapy. Further studies should investigate more appropriate r-IL-2 schedules and the possibilities offered by better antigen recognition and activated effector cells.

Treatment of chronic myelogenous leukemia with interleukin-2: a phase II study in 21 patients.

We designed a phase II study to assess the activity of recombinant interleukin-2 (rIL-2) in patients with chronic myelogenous leukemia (CML). Study population included 11 patients in the chronic phase of CML (6 in hematologic remission and 5 with active disease), 6 patients in the accelerated phase, and 4 in blastic phase of CML. Patients received three 5-day cycles administrated every other week. rIL-2 was given as intravenous bolus infusions of 8 x 10(6) IU/m2 three times a day during cycle 1 and twice a day during cycles 2 and 3. Response to rIL-2 was assessed on day 45. No hematologic response was achieved in the patients with evaluable disease. One patient in hematologic remission with rIL-2 achieved a major response (from 72% to 9% Ph+ metaphases), and two patients had some degree of reduction of Ph+ metaphases. Responses were short-lived (< 6 months), but two of these three patients achieved a new cytogenetic response with interferon given post-rIL-2. A significant immune activation was achieved with rIL-2 including a marked increase in CD3+/CD25+ cells, CD56+ cells, and in natural killer/lymphokine activated killer cell cytotoxic activity. These results confirm preclinical studies, which showed that IL-2 has antileukemic activity in CML. However, the responses observed were short lived and restricted to a subgroup of patients with low disease burden. This invites further studies testing its impact in situations of minimal disease or in combination with other cytokines.

A natural cytotoxic T cell response in a spontaneously regressing human melanoma targets a neoantigen resulting from a somatic point mutation.

We have studied a case of human primary melanoma displaying the classical signs of a spontaneous regression in order to characterize potentially efficient anti-tumor T cell responses. In a previous series of experiments a unique TCR Vbeta16+ T cell was shown to be highly expanded at the tumor site. The corresponding clone was isolated in vitro and found to be a CD8+ cytotoxic T lymphocyte with a strong and selective cytolytic activity against the autologous tumor cell line. Here, we demonstrate that this predominant Vbeta16+ tumor-infiltrating lymphocyte recognizes a peptide encoded by a novel unconventional myosin class I gene. This peptide includes a mutation due to a single nucleotide substitution. The resulting Glu-->Lys replacement at position 911 of the coding sequence is critical to generate the recognized T cell epitope. These experiments demonstrate the existence of a natural tumor-specific cytolytic T cell response in a primary regressing human melanoma lesion.

A MAGE-6-encoded peptide is recognized by expanded lymphocytes infiltrating a spontaneously regressing human primary melanoma lesion.

In recent years, experiments based on the in vitro stimulation of either autologous peripheral blood lymphocytes or tumor-infiltrating lymphocytes with melanoma cells have shown that distinct members of the large MAGE gene family encode tumor-associated antigenic peptides. However, little is still known about natural anti-MAGE responses in vivo. We have studied a case of spontaneously regressing human melanoma, hypothesizing that in this unique situation, the host immune system had developed an efficient cytotoxic T lymphocyte (CTL) response against the cancer cells. Amongst the dense tumor infiltrate, certain clonal populations of T cells were shown to be amplified, thereby suggesting that an antigen-driven selection had occurred at the tumor site. One of the expanded tumor-infiltrating lymphocytes was shown to be a Vbeta13+ CD8+ CTL displaying a strong and selective cytotoxic activity against the autologous melanoma cells. Here we show that this cytotoxic T cell clone recognizes a MAGE-6-encoded peptide. MAGE-6 is therefore the fourth gene of the MAGE family shown to encode antigenic peptide recognized by T cells. Together, these data provide further evidence that T cell responses against MAGE antigens may naturally develop in vivo.

A phase II study of interleukin-2 in 49 patients with relapsed or refractory acute leukemia.

In this report we present the results of a multicenter phase II study of high-dose recombinant Interleukin-2 (rIL-2) in patients with refractory or relapsed acute leukemia. Forty-nine patients with acute myeloid leukemia (AML: 30 patients) or acute lymphoblastic leukemia (ALL: 19 patients) were included. Median age was 30 years (range: 4-71). Four patients were treated for primary refractory disease and 45 for relapsed disease (16 patients > 2nd relapse). Twenty-four patients (49%) had previously received bone marrow transplantation (allogeneic: 5, autologous: 19). Patients were scheduled to receive three 5-day cycles of rIL-2 given every other week. rIL-2 was administered as bolus I.V. infusion of 8 x 10(6) UI/m2 every 8 hours during cycle I and every 12 hours during cycles 2 and 3. Patients received a mean of 76% of rIL-2 planned dose. Main toxicity was hematologic (grade IV thrombopenia: 84%). Hemodynamic and metabolic toxicities lead to treatment discontinuation in 10 patients (20%). Strong immune activation was achieved including a significant increase in activated T-cells and Lymphokine-Activating-Killer cell (LAK) activity. Twenty-seven out of 30 AML patients could be evaluated for response: 2(7%) achieved complete remission (CR) which lasted 3 and 4 months. No response was observed in the 18 assessable ALL patients, most of whom (77 %) presented absolute drug resistance. These results show that this high dose rIL-2 regimen induces significant biological effects and provides some anti-leukemic activity in patients with advanced leukemia. Considering the severe toxicity observed and the limited remission rate achieved here, rIL-2 does not appear to be a valuable therapeutic option for such patients. However, the undoubted anti-leukemic activity of this cytokine invites further investigation especially in the minimal residual disease situation.

Analysis of T-cell immune response in renal cell carcinoma: polarization to type 1-like differentiation pattern, clonal T-cell expansion and tumor-specific cytotoxicity.

We assessed the naturally occurring T-cell immune response in primary renal cell carcinoma (RCC) tumors from 12 unselected patients. A predominance of CD3+ T-cell receptor (TCR)alpha/beta+ T cells was observed in tumor-infiltrating lymphocytes (TILs), in contrast with peripheral blood lymphopenia found in some patients. Activation antigen expression on TILs revealed an imbalance in the activation status, with a significant percentage of CD69+ and HLA-DR+ and a low percentage of CD25+ and CD71+ TILs. The lymphocyte activation gene-3 (LAG-3) was detected in some TIL subpopulations and especially in one patient in whom TILs were predominantly TCR alpha/beta+CD8+DR+LAG-3+. In addition, we found that RCC TILs are polarized to a global type 1-like (Th1/Tc1) differentiation pattern (strong secretion of interferon-gamma and interleukin-2 (IL-2) following CD3/TCR crosslinking) but are under the influence of the down-modulatory cytokines IL-6 (secreted by tumor cells) and IL-10, within the tumor microenvironment. In 3 of 5 patients, clonal T-cell expansion at the tumor site was found for several Vbeta specificities, suggesting that in situ stimulation of specific clonotypes in response to potential tumor antigens is a frequent event in RCC. Furthermore, in one patient, selective intratumor amplification of a Vbeta1 subpopulation (5% of TCR alpha/beta+ cells) corresponding to 2 distinct Vbeta1-Jbeta1.6 and Vbeta1-Jbeta2.3 tumor-specific MHC class I-restricted cytotoxic T lymphocytes supports the view that discrete T-cell subsets contribute readily to in situ immunosurveillance.

In situ T-cell responses in a primary regressive melanoma and subsequent metastases: a comparative analysis.

In an earlier study of the immune response in a patient with a cutaneous primary regressive melanoma, a T-cell-receptor diversity analysis demonstrated in situ amplification of certain lymphocytes. Two of them could be cloned and characterized as CD8+ HLA-class-l-restricted CTL with strong selective anti-tumor activity. Following a disease-free period of 3 years, the patient developed a gastric metastasis and subsequently (after an additional year) a metastasis in one axillary lymph node. Melanoma cell lines derived from the 2 secondary lesions have been established here. It was found that these metastatic cells have maintained expression of both HLA-class-I molecules and the peptidic antigen(s) recognized by the 2 clones amplified at the primary site. However, the corresponding T lymphocytes were either undetectable or poorly represented both in the gastric and in the axillary lesions. These results suggest that substantial alterations in the quality of T-cell infiltrates occurred during melanoma progression, despite an apparent stability in presentation of tumor-associated antigen(s) which initially triggered a positive rejection response.

MHC-dependent cytolysis of autologous tumor cells by lymphocytes infiltrating urothelial carcinomas.

Tumor-infiltrating lymphocytes (TIL) were grown from 23 urothelial carcinomas. Phenotyping analysis showed that the TIL cultures were mainly CD3+. Although CD4+ and CD8+ T-cell sub-sets were grown in culture, CD4+ T-cell sub-sets predominated over CD8+ T cells. Immunohistochemical studies performed on 5 tumor specimens confirmed this observation, and indicated that CD4+ T cells surrounded the tumor islets, whereas CD8+ T lymphocytes were localized among the tumor cells. Five short-term carcinoma cell lines established from these urothelial tumors were used as target cells in cytolysis assays in order to investigate the functional anti-tumor activity of autologous TIL. TIL from 4/5 tumors were lytic and 3 TIL lines displayed MHC-class-I-dependent cytotoxicity directed against autologous tumor cells. CD4+ T-cell-depletion experiments performed on TIL line 07 confirmed that CD8+ MHC-class-I-dependent CTL were the predominant effectors. Finally, experiments performed on 6 allogeneic urothelial-cancer cell lines matched for HLA-class-I molecules showed that TIL07 exhibited selective lytic activity toward tumor 07. These data indicate that CD8+ MHC-class-I-dependent CTL present in urothelial carcinomas are functional and may participate in the anti-tumor immune response.

The use of a sequential high dose recombinant interleukin 2 regimen after autologous bone marrow transplantation does not improve the disease free survival of patients with acute leukemia transplanted in first complete remission.

We report the outcome of 50 consecutive patients with CR1 acute leukemia (AML = 22; ALL = 28) treated with autologous BMT, after cyclophosphamide and TBI, followed with a sequential high dose rIL2 regimen. rIL-2 (RU 49637 from Roussel-Uclaf, Romainville, France) was started after hematological reconstitution an average of 72 +/- 22 days post transplant. The schedule consisted of a continuous infusion over 5 cycles (Cycle 1: 5 days starting on day 1; cycle 2-5: 2 days starting on day 15, 29, 43 and 57). Patients were treated at 4 different dosages (12 (N = 40), 16 (N = 3), 20 (N = 2), 24 (N = 5) x 10(6) IU/m2/day). Toxicities were mainly related to capillary leak syndrome and thrombocytopenia. Patients received an average of 122 +/- 49 10(6) IU/m2. Two patients with AML died from toxicity. rIL-2 infusion was associated with very a high level of immune stimu-lation of both T-cells (P < 0.05) and natural killer (NK) cells (P < 0.05) and associated cytolytic functions (P < 0.05). With a minimal and median follow-up of 21 and 46 months, 3 year leukemia free survival is 41 +/- 6% overall, 39 +/- 10% and 43 +/- 8% for AML and ALL respectively. Relapse probabilities at 3 years are 59 +/- 11% for AML and 57 +/- 8% for ALL. We conclude that this short infusion of rIL-2 over 2 months, resulting in an increased immune stimulation, is not associated with a better leukemic control for patients with acute leukemia transplanted early after reaching first complete remission.

T-cell receptor repertoire in neuroblastoma patients.

Spontaneous regression of widespread lesions is a characteristic feature of neuroblastoma. One may postulate that the immune response contributes to these clinical regressions. Accordingly, we studied the T-cell receptor (TCR) repertoire of tumor-infiltrating lymphocytes in eight neuroblastoma tumors. The expression of 29 V alpha and 24 V beta gene segment subfamily specificities was analyzed by PCR and compared by computerized densitometry of Southern blots to values obtained in the blood. Overall, the TCR repertoire of these eight patients was diverse, with virtually all V alpha and V beta specificities expressed. Nonetheless, four of these patients showed V beta 2 gene segment subfamily overexpression in the tumor corresponding to local expansion of polyclonal T-cell subpopulations. In one patient, this expansion could be due to local secretion of superantigenic activity, as suggested by the specific stimulation of murine T cells expressing a human V beta 2 chain by supernatant of the corresponding neuroblastoma cell line. In addition, high-resolution analysis of the TCR beta transcript complementarity-determining region 3 sizes identified three patients (of six studied) with marked clonal T-cell expansion in the tumor not seen in the blood. The specific expression of several dominant clono-types in the tumor may be related to the recognition of neuroblastoma-specific antigens in these patients. Together, these results on the TCR repertoire expressed in vivo may lead to the characterization of putative immune response mechanisms (i.e., antigen- or superantigen-driven stimulation) which participate in tumor regression.

Dihydroorotate dehydrogenase is a high affinity binding protein for A77 1726 and mediator of a range of biological effects of the immunomodulatory compound.

A protein with high affinity (Kd 12 nM) for the immunomodulatory compound A77 1726 has been isolated from mouse spleen and identified as the mitochondrial enzyme dihydroorotate dehydrogenase (EC 1.3.3.1). The purified protein had a pI 9.6-9.8 and a subunit Mr of 43,000. Peptides derived from the mouse protein displayed high microsequence similarity to human and rat dihydroorotate dehydrogenase with, respectively, 35 and 39 out of 43 identified amino acids identical. Dihydroorotate dehydrogenase catalyzes the fourth step in de novo pyrimidine biosynthesis. The in vitro antiproliferative effects of A77 1726 are mediated by enzyme inhibition and can be overcome by addition of exogenous uridine. The rank order of potency of A77 1726 and its analogues in binding or enzyme inhibition was similar to that for inhibition of the mouse delayed type hypersensitivity response. It is proposed that inhibition of dihydroorotate dehydrogenase is an in vivo mechanism of action of the A77 1726 class of compounds. This was confirmed using uridine to counteract inhibition of the murine acute graft versus host response.

[Treatment of metastatic kidney neoplasms with a new interleukin 2 protocol: The experience of the Gustave-Roussy Institute]. / Traitement du cancer du rein métastatique avec un nouveau schéma d'interleukine 2: expérience de l'institut Gustave-Roussy.

Treatment of metastatic renal cell carcinoma with interleukin 2 (IL2) remains controversial despite the authorization from the French government for IL2 with the West schedule in this disease. We report herein the study of the Institute Gustave-Roussy of 73 patients, who received from 1989 to 1991 a new schedule of high dose IL2. Seventy three patients received high dose IL2 according to the following schedule: IL2 by continuous infusion at 24.10(6) IU/m2/d, on 2 consecutive days per week, during 5 weeks. This treatment was associated in the first 33 patients with gamma interferon at a dose of 5.10(6) IU/m2/d subcutaneously the days of IL2 infusion, during the 5 weeks of therapy. Immunotherapy was further continued in responding patients, either as an association of IL2 and LANAK (lymphokine-activated natural killer) cells, or as IL2 alone. Finally, when possible, surgery was performed on residual masses. Twenty five percent of objective responses (PR + CR) have been observed. Moreover, 12.3% CR has been obtained after the overall therapy. The global mean survival is 15 months, with a mean survival of 8, 18 and 24+ months depending on the status of the disease (progressive, stable or responding) after initial treatment with IL2. Tolerance of this schedule was good with an actual received dose of 90% of the planned doses, and patients could leave the hospital within 2 hours after the end of IL2 in 87% of the cycles. No toxic death was observed. Among the parameters observed for correlation with the clinical response, only performance status and level of sTNF-alpha R were significantly associated with the response.

[Comparative analysis of the immune response in a case of primary regressive melanoma followed by gastric metastasis]. / Analyse comparative de la réponse immunitaire dans un cas de mélanome primitif régressif suivi d'une métastase gastrique.

The theory of immunosurveillance against cancer has been an extensively debated question over the last decades. Multiple indirect arguments have supported the view that the immune system may control, at least in certain cases, malignant cell growth while direct demonstration is still lacking in the human. In an attempt to address this issue, we have selected a study model, namely spontaneously regressive melanoma. A primary cutaneous lesion was investigated. T cell repertoire analysis showed the in situ amplification of at least two tumor infiltrating lymphocyte (TILs) expressing the V beta 13 and V beta 16 variable TCR gene segments respectively. The two clones were precisely characterized by sequence of the TCR beta chain junctional region. Further functional study demonstrated that both lymphocytes displayed a selective cytotoxic activity against autologous tumor cells. The V beta 16+ cells, predominant in vivo, were shown to be closely opposed to the melanoma cells. Together, these studies demonstrated the existence of a local adaptative immune response associated to tumor regression, thus strongly supporting the validity of the immunosurveillance concept. A gastric metastasis which occurred three years after the primary lesion has been studied here. Overexpression of the V beta 13 and V beta 16 TCR segments was no longer detected by direct PCR analysis in situ. Sequencing transcripts from V beta 13+ and V beta 16+ TILs confirmed that the two CTLs, identified in the primary lesions, were not represented with high frequency. The V beta 13+ cell was however shown to be present while the V beta 16+ CTL was not detected. Yet, characterization of a tumor cell line derived from the gastric site indicated that the peptidic antigen(s) which induced the initially successfully immune response was still expressed. The present data illustrate that it has become possible to perform very precise analysis of local immune responses during cancer development. Such an improvement together with the recently initiated molecular characterization of tumor associated peptidic antigens, should provide the basis for improved strategies of cancer immunotherapy.

T-cell repertoires in healthy and diseased human tissues analysed by T-cell receptor beta-chain CDR3 size determination: evidence for oligoclonal expansions in tumours and inflammatory diseases.

Many examples of oligoclonal T-cell expansion in infiltrated diseased tissues have been reported. However, it remains to be established whether such observations can be generalized and to what extent oligoclonal patterns obtained after in vitro culture of T-cell infiltrates reflect in vivo situations. Using new high resolution analysis which requires no in vitro cellular expansion, we detected such oligoclonal T-cell expansions in 7/7 melanoma tumour biopsies, 3/3 biopsies of inflammatory skin during acute graft versus host disease (aGVHD) after allogeneic bone marrow transplantation (alloBMT) and 7/7 synovial membranes from patients with rheumatoid arthritis. Thus, oligoclonal T-cell expansions are readily observed when a sufficiently sensitive detection method is used, suggesting that similar expansions are the rule among T-cell infiltrates in different diseases. This observation and the monitoring of the in vivo evolution of such expansion during the course of the disease and during in vitro culture should have important clinical implications.

Lymphocyte-activation gene 3/major histocompatibility complex class II interaction modulates the antigenic response of CD4+ T lymphocytes.

The activation requirements for antigen-dependent proliferation of CD4+ T cells are well documented, while the events leading to the inactivation phase are poorly understood. Here, we tested the hypothesis that the lymphocyte-activation gene 3 (LAG-3), a second major histocompatibility complex (MHC) class II ligand, plays a regulatory role in CD4+ T lymphocyte activation. CD4+ class II-restricted T cell clones were stimulated by their relevant antigen (hemagglutinin peptide or diphteria toxoid) and antigen-presenting cells with or without anti-LAG-3 monoclonal antibody (mAb). Kinetic studies were performed to monitor different activation parameters, including the measurement of thymidine incorporation, expression of activation antigens and cytokine secretion. Results showed that the time course from the initial time points up to the peak time point was not modified in the presence of anti-LAG-3 mAb. However, addition of these antibodies, either as whole IgG or as Fab fragments, led to increased thymidine incorporation values for late time points and, hence, to a shift in the decreasing proliferation curve. We also showed that expression of activation antigens, such as CD25, was higher in the presence of anti-LAG-3 mAb, and that cytokine concentrations, i.e. of interferon-gamma or interleukin-4, were higher in the corresponding culture supernatants. In addition, we tested whether the effects of anti-LAG-3 mAb were limited to antigen-dependent, MHC class II-restricted responses. The proliferative responses of CD4+ T cell clones following stimulation with either interleukin-2, mitogens, a combination of anti-CD2 mAb, immobilized anti-CD3 or anti-T cell receptor mAb were not altered by anti-LAG-3 mAb. The allogeneic proliferative response of a CD8+ T cell clone was also not affected. Overall, the present analysis reveals a modulating effect of anti-LAG-3 mAb, mediated specifically on antigen-dependent, MHC class II-restricted responses of CD4+ T cell lines. These results support the view that LAG-3/MHC class II interaction down-regulates antigen-dependent stimulation of CD4+ T lymphocytes.

T cell repertoire in patients with B chronic lymphocytic leukemia. Evidence for multiple in vivo T cell clonal expansions.

To characterize circulating T cell subpopulations in B chronic lymphocytic leukemia patients, TCR V alpha and V beta gene-segment use was analyzed by PCR using a panel of V gene-segment subfamily-specific oligonucleotide primers (V alpha 1-29/V beta 1-24). Virtually all V alpha and V beta subfamily specificities were expressed in these patients (nine stage A and four stage C), and the mean values obtained for each specificity were similar to those of a group of 13 healthy donors. Nonetheless, individual analysis revealed that unique V alpha or V beta gene-segment transcripts were overrepresented in patients compared with the control group. Overrepresentation of some TCR V beta chains was also detected by cytofluorometric analysis using a panel of 18 anti-V beta-specific mAbs. To further characterize these T cell subpopulations, we sequenced five different V beta-C beta PCR products in two selected stage A patients and found highly predominant recurrent transcripts in each of the five V beta specificities (50% to 100% of the analyzed sequences with identical V(D)J regions). These results were confirmed on bulk cDNA (i.e., without cloning) and extended to other V beta specificities (up to nine clonal expansions of 24 V beta specificities in one patient) and two other patients using a PCR-based method that determines V(D)J junction size patterns. Finally, it was observed that a V beta 19+ T cell subpopulation was clonally expanded in one patient to up to 30% of circulating T cells. This V beta 19+ CD8+ T cell clone was shown to specifically recognize the autologous tumor cells in vitro, as determined in cytokine release assays. Together, these results support the view that multiple expansions of unique T cell clones may derive in vivo from B chronic lymphocytic leukemia tumor-associated Ag stimulation.

Influence of human leukocyte antigen genes on TCR V gene segment frequencies.

Human leukocyte antigen (HLA)-dependent selection mechanisms exerted during thymic maturation are supposed to be main contributing factors to the genetic predetermination of the TCR repertoire and may have a detectable effect on adult peripheral blood lymphocyte V segment frequencies. Here, we analyzed whether polymorphic or non-polymorphic HLA determinants are associated with selected expression of some V gene segment specificities. We first examined the reactivity of 17 V segment specific mAb on purified CD4+ and CD8+ cell fractions in 10 unrelated people. We found a significant overexpression of only three V segment products (V beta 2, V beta 5.1 and V beta 6.7) in CD4+ and none in CD8+ cell fractions in most individuals. Skewing of certain V beta segments by non-polymorphic HLA determinants (i.e. class II for CD4+ and class I for CD8+ cells) is therefore more limited (3/17) than previously thought. Considering the effects of polymorphic HLA determinants, we compared TCR V segment frequencies in HLA-identical siblings to sibling pairs who differ at one or both HLA haplotypes, using 13 V beta specific mAb. In pairwise comparisons, we found that the HLA complex had no detectable effect on TCR repertoire in five large families with multiple siblings. Together, these observations suggest that HLA-predicted selection mechanisms exerted during thymic maturation might not have a predominant influence shaping the TCR repertoire of normal adults.

In vivo T-cell clonal amplification at time of acute graft-versus-host disease.

In a series of patients transplanted with HLA-matched allogeneic bone marrow grafts (alloBMT), we previously showed that a few T-cell receptor (TCR) V alpha and V beta gene segment transcripts were overexpressed in skin compared with blood at the time of acute graft-versus-host disease (aGVHD). Here, in one selected patient with overexpressed V beta 16 and V alpha 11 transcripts in skin, we analyzed the junctional variability of these transcripts in donor blood, patient blood, and skin collected at aGVHD onset. A unique junctional region sequence accounted for 81% of in frame V beta 16 transcripts (13 of 16) in skin and 59% (13 of 22) in patient blood. Similarly, two recurrent junctional region sequences were found in skin V alpha 11 transcripts, one accounting for 66% (21 of 32) and the other for 16% (5 of 32). These recurrences were also found in patient blood (36% and 15% of V alpha 11 transcripts, respectively). To extend our analysis, a polymerase chain reaction (PCR)-based method was used to precisely determine TCR beta transcript length in run-off reactions using uncloned bulk cDNA samples. All V beta-C beta PCR products analyzed in donor blood, as well as the majority of those analyzed in patient blood, included transcripts with highly diverse junctional region sizes. As expected from the sequence data, most V beta 16-C beta PCR products in skin and patient blood were of the same size (ie, same junctional region). In addition, V beta 3, V beta 5, and V beta 17 transcripts in skin were shown to display highly restricted size variability. The clonality of the V beta 16-C beta and V beta 17-C beta transcripts was further supported by the results of run-off reactions using 13 J beta specific primers. We have identified several recurrent TCR transcripts in skin, some of them also present in patient blood. These data support the view that several T-cell subpopulations are clonally expanded in vivo at the time of aGVHD onset in this case of related HLA-matched alloBMT.