Pilot Study of Radiation-Targeted Donor Lymphocyte Infusion for Cancer Progression after Allogeneic Hematopoietic Stem Cell Transplantation

Immune escape from graft-versus-tumor (GVT) contributes to relapse after allogeneic stem cell transplantation (SCT). High-dose radiation (XRT) mobilizes a tumor-specific immune response following direct damage to the tumor and microenvironment. Radiation's abscopal effect is also immune mediated. We hypothesized that local radiation of relapsed tumor would rekindle a GVT effect and yield abscopal tumor responses.

We tested the safety and systemic effects of single-fraction XRT followed by donor lymphocyte infusion (DLI) in patients with lymphoid malignancy and refractory relapse after SCT. Eligible subjects (Table) had tumor progression following donor engraftment, withdrawal of immune suppression and/or DLI, and XRT-amenable tumor. Subjects received 8 Gy XRT to 1 or more tumors; all but 1 (with chronic GVHD) received DLI 24 hours later. Clinical monitoring included marrow function, GVHD, XRT toxicity and tumor response. Correlative monitoring for systemic immunologic and GVT effects included FDG-PET imaging, serial peripheral blood sampling, and biopsy of radiated and nonirradiated tumor. Immune effects were characterized by flow cytometry and gene expression analysis.

Local XRT toxicity of Grade 3 or less presented as expected in all subjects. Grade 4 neutropenia (2) was transient and GCSF-responsive; GVHD flare was not observed. Unexpected toxicity was observed in 1 subject who developed Grade 4 diffuse alveolar hemorrhage and Grade 3 cardiomyopathy that responded to therapy.

Sustained response in radiated sites was observed in 6 of 8 evaluable subjects (nonevaluable: 1 MM bony lesion; 1 subject received further Rx). RECIST responses outside the radiation field included 2 PR and 5 SD. 2 subjects with SD had regression after initial tumor flare. 3 of 4 regressions were transient, with progression by 6 months; 3 subjects demonstrated progression. A single subject who received radiation alone demonstrated systemic tumor response as well as improvement in chronic GVHD (sclerotic skin).

By FDG-PET, all subjects' radiated tumors showed substantially decreased uptake at D28; SUV changes were variable at D7. Changes in non-XRT tumors demonstrated far greater variability, even within single subjects.

Consistent with systemic immune activation resulting from radiation-DLI therapy, within 1 week, peripheral blood T cell proliferation (%Ki-67+) increased significantly in CD4 and CD8 T cells, as well as FoxP3+ Tregs (Wilcoxon paired nonparametric analyses; p =.004, .004 and .01, respectively). A trend (p=.078) toward increased frequency of Th1 effector CD4 cells (T-bet+CD28-) was also observed.

Comparison of gene expression in nonirradiated (abscopal) tumor in 4 patients collected pre and 1 month post XRT was consistent with an IFN-mediated response to cell damage. Genes were upregulated for receptors for nuclear materials released by damaged cells (TLR2, TLR4, CLEC4E), high- and low- affinity FcgR (FCGR1A and FCGR2A) used for phagocytosis, and an annexin receptor for phagocytic cells (FPR2). The inflammasome component genes NLRP3 and CASP1 were upregulated, as was IFIT1, a Type 1 IFN inducible gene.

After SCT, XRT results in local tumor regression; abscopal responses were also observed, with systemic immune responses suggested by T cell proliferation in the peripheral blood and upregulation of tissue damage receptors and IFN-inducible genes in nonirradiated tumor. Systemic tumor responses were delayed, with peak regression occurring 3-4 months after XRT, consistent with an immune-mediated effect. GVHD was not observed; significant, reversible pulmonary and marrow toxicities were infrequent. Transient abscopal responses in these DLI-refractory patients suggest plausible synergy between radiation and allogeneic GVT effects. Further investigation into posttransplant immunomodulatory radiation is warranted.

Funding: Intramural NCI/CCR and NCI Contract No. HHSN261200800001E