Adenovector-mediated gene delivery of interleukin-2 in metastatic breast cancer and melanoma: results of a phase 1 clinical trial.

We conducted a phase 1 trial of direct injection of an E1, E3-deleted adenovirus encoding interleukin-2 (AdCAIL-2) into subcutaneous deposits of melanoma or breast cancer. Twenty-three patients were injected at seven dose levels (10(7)-10(10) p.f.u). Local inflammation was observed at the site of injection in 60% of patients, but side-effects were otherwise minor. Incomplete local tumor regression occurred at the site of injection in 24% of patients, but no conventional clinical responses were seen. Circulating CD4 and CD8 counts fell significantly 24 h after injection. Post-injection biopsies demonstrated tumor necrosis and lymphocytic infiltration with the predominant tumor-infiltrating cells both CD3- and CD8-positive. Vector-derived sequences were detected in 14 of 18 biopsies examined 7 days after injection and vector-derived hIL-2 mRNA was detected in 80% of 7-day biopsies processed after injection of 10(8) p.f.u. of AdCAIL-2 or higher. While IL-2 was detectable by ELISA in tumor biopsies at 48 h, no protein was detectable in injected tumors after 7 days and no circulating IL-2 was detectable at any time-point. No Ad5E1 sequences were detected either before or after injection indicating absence of replication-competent virus or endogenous E1-like sequence; furthermore, only rare vector shedding was detected. Anti-adenovirus and neutralizing antibody titers were elevated 1 month after injection in all patients. This trial therefore confirms the safety of use of adenoviral vectors for gene delivery in humans and demonstrates successful transgene expression even in the face of pre-existing immunity to adenovirus.

Erythropoiesis: Splenic Immunoglobulin Variable Region Genes Encoding Red Blood Cell Binding Fab Fragments in Autoimmune Hemolytic Anemia.

An absolute requirement for the V(H(4-34) ) immunoglobulin (Ig) variable (V) region heavy chain (V(H) ) gene has been demonstrated in pathogenic cold agglutinin autoantibodies. Investigation of IgG binding anti-Rhesus (Rh) alloantibodies provides further evidence of V gene restriction in red blood cell (RBC) binding antibodies and demonstrates that the V(H(4-34) ) gene used to form cold agglutinins may also encode RBC antibodies of varied specificities. We reasoned that a similar V gene restriction may be evident in the gene segments encoding IgG anti-RBC autoantibodies mediating autoimmune hemolytic anemia (AIHA). To further examine this question IgG Fab fragment phage display libraries were constructed from the spleen of a patient with AIHA. The index autoantibody appeared to have incomplete anti-C specificity and bound all panel RBCs except Rh null. The Fab fragment phage display libraries were therefore panned twice on CDE/CDe RBCs and binding phage were eluted. Binding of the phage displayed Fab fragments to RBCs was confirmed by immunoflourescence and flow cytometry. Specificity was confirmed by the absence of binding to Rh null cells, murine RBCs and to human peripheral blood lymphocytes. Molecular analysis of Ig V genes encoding the pan RBC binding Fab fragments revealed a relative V(H) gene restriction and evidence of somatic mutation. The V(H(3) ) family member V(H(26) ) was prominent in RBC binding Fabs. The V(H(3) ) family member hV3005 and the V(H(4) ) family DP-65 gene segments also encoded RBC binding Fabs. The J(H(4) ) gene segment was present in all binding clones. Varied kappa and lambda light chain (V(L) ) genes were identified by sequencing and no single light chain was prevalent. Three of the ten V(L) and two of the three V(H) identified by sequencing appeared to derive from germline genes previously noted to have RBC binding specificity. We conclude that splenic Ig V genes can encode pan RBC binding antibodies with specificities similar to autoantibodies found in AIHA and that V(H) gene segment utilization by these antibodies is derived from a limited pool of somatically mutated V(H) gene segments.

Molecular and serological examination of the relationship of human herpesvirus 8 to multiple myeloma: orf 26 sequences in bone marrow stroma are not restricted to myeloma patients and other regions of the genome are not detected.

Human herpesvirus 8 (HHV-8) genomic sequences were recently detected by polymerase chain reaction (PCR) and in situ hybridization in bone marrow stromal cells grown from multiple myeloma (MM) patients, but not in cells from control subjects (Rettig et al, Science 276:1851, 1997). We sought to confirm these observations in our own group of MM patients (n = 30). DNA was extracted from adherent stromal cells grown under varying conditions and assayed for HHV-8 sequence using PCR to amplify the orf 26 (KS330) sequence (Chang et al, Science 266:1865, 1997), as initially reported. Samples from human control subjects (n = 25) were concurrently extracted and analyzed. After 30 cycles of amplification, we did not detect any positive samples. In a more sensitive nested PCR, samples from 18 of 30 (60%) MM patients were positive, at about the limit of detection, but orf 26 sequence was also amplified from 11 of 25 (44%) human control samples. However, PCR amplification from other regions of the viral genome (orf 72 and orf 75) was uniformly negative for all MM and control samples, despite equivalent sensitivity. Additionally, all sera from MM patients were negative for HHV-8 IgG by immunofluorescence. Our data do not support a role of HHV-8 in the etiology of MM but may suggest the presence of a related (KS330-containing) virus in MM patients and in some control subjects. This is a US government work. There are no restrictions on its use.

In vitro maintenance and retroviral transduction of human myeloma cells in long-term marrow cultures.

One objective of clinical gene marking trials in multiple myeloma (MM) is to determine the extent to which relapse after stem cell transplant is attributable to contamination of the autograft with myeloma cells. A requirement in these studies is ex vivo genetic marking of malignant cells present in autografts which are derived from patients exposed to significant prior chemotherapy. We evaluated gene marking of cloonogenic myeloma cells in marrow aspirates from 14 patients with MM. To effect gene transfer we utilized a long-term marrow culture (LTMC) system previously shown to facilitate gene transfer into a spectrum of hematopoietic progenitor and stem cells. Transduction of cells in LTMC was performed by multiple supernatant exposure. At LTMC initiation and after 21 days of culture malignant cells were assessed by morphology, flow cytometry, and polymerase chain reaction (PCR). The mean number of day 21 LTMC adherent layer-derived granulocyte/macrophage progenitors as a percentage of the original inoculum was within the normal range for this technique. The efficiency of transduction of normal hematopoietic progenitors as determined by the number of colonies positive for proviral DNA by PCR, G418 resistance, and X-gal staining was also within the expected range; 65%, 44% and 23%, respectively. Thus, there was no evidence that prior chemotherapy exposure or malignant cell contamination compromised cell survival or gene transfer efficiency in LTMC. All patients retained plasma cells in LTMCs for the duration of the 21-day culture period. Molecular analysis confirmed the persistence of clonal IgVH gene rearrangements in day 21 LTMC-derived DNA from 6 of 12 informative patients (50%). PCR using allele-specific primers when available confirmed the specificity of IgVH rearrangements for the myeloma clone. In 2 of the 14 patients, expansion of clonogenic cells was demonstrated in LTMC. In both cases there was strong evidence for transfer of reporter genes (neo and LacZ) into the myeloma clone: morphologically abnormal G418-resistant colonies demonstrated intense staining for beta-galactosidase, and cytospin preparations showed 100% plasma cells with monoclonal heavy and light chain restriction. In one patient, individual colonies positive for beta-galactosidase bore a cytogenetic abnormality characteristic of the patient's myeloma clone. PCR of DNA from pooled plasma cell colonies using tumor-specific CDR3 primers was positive. Our results demonstrate the maintenance of myeloma cells in vitro for up to 21 days in LTMC. They further illustrate that these cells can be genetically marked using transduction protocols currently being tested in clinical trials of hematopoietic cell gene transfer.