Interleukin-2 May Enhance Gene Therapy of Melanoma

SAN FRANCISCO—Investigations into the cellular basis of the anticancer activity of recombinant interleukin-2 (IL-2, Proleukin) may lead to immunization against some cancers. That was the prospect suggested by Steven A. Rosenberg, MD, head of the NCI’s Surgery Branch and professor of surgery at George Washington University School of Medicine and Health Sciences, and the Uniformed Services University of the Health Sciences, Bethesda, Md.

Animal models and phase I human trials suggest that IL-2 and genetic manipulation may produce therapeutically significant immunization against cancer antigens. The work could lead to therapeutic agents to treat or prevent specific cancers.

About 20% Respond to IL-2

About 20% of patients with metastatic melanoma or renal cell carcinoma who are treated with high-dose IL-2 respond to therapy. But unlike the responses observed following treatment with traditional chemotherapeutic agents, IL-2 re-sponses are remarkably durable.

About 10% of patients treated with IL-2 show complete remission. Of these complete responders, about 75% have no disease recurrence. Some of the complete remissions have lasted 10 years and longer, even though the traditional prognosis for metastatic melanoma and renal cell carcinoma is extremely poor.

“If you’re a complete responder at five years,” Dr. Rosenberg said at the Proleukin First International Congress, sponsored by Chiron, “there is a great likelihood that you have been cured of your metastatic disease.”

The Cellular Basis

Interleukin-2 has no direct effect on cancer cells. It acts entirely by modulating the immune system. This fact led Dr. Rosenberg to investigate the cellular basis of IL-2’s antitumor effects.

His group of researchers identified tumor-infiltrating lymphocytes (TILs) that recognize unique cancer antigens on human (and murine) lymphomas as well as melanomas, breast cancers, and colon cancers.

Clinical trials of TILs plus IL-2 produced a 36% objective remission rate in patients with metastatic melanoma, Dr. Rosenberg said. These initial results encouraged their efforts to genetically modify TILs in order to boost their therapeutic effect.

Tumor-infiltrating lymphocytes were initially transfected with the gene for neomycin resistance and administered to patients with advanced cancer in order to verify selective activity. The labeled TILs were detected first in the tumor and later in circulating serum.

TILs transfected with the tumor necrosis factor (TNF) gene have been used to treat 11 patients with advanced melanoma; results were not released at the Congress.

Tumor-infiltrating lymphocytes have also been transfect-ed with chimeric T-cell receptors combining the constant region of the T-cell receptor and the variable region of a tumor-specific monoclonal antibody, Dr. Rosenberg said. The altered TILs bind selectively to the cancer recognized by the monoclonal antibody to concentrate TIL-mediated lysis at the most therapeutically significant site.

The next step was to isolate antigens specifically associated with cancer rejection in patients. TILs from metastatic melanoma patients that selectively recognized autologous cancer antigens in vitro were identified, administered to the patient, and screened for activity against the melanoma in vivo, then cloned in culture.

The genes that encode the cancer antigens recognized by TILs were cloned, and patients were immunized with the genes or the gene products to determine if they could produce tumor regression.

Six Genes Identified

So far, six genes encoding tumor antigens have been identified, as well as the immunodominant peptides associated with each of the genes. The most common antigens are gp100 and a previously unknown protein dubbed MART-1. More than 90% of melanomas express at least one of these antigens, most often MART-1.

Phase I immunotherapy trials using TILs that had been altered to recognize MART-1 and gp100 in metastatic melanoma patients began in early 1995.

Trials of recombinant vaccinia, adenovirus, and fowlpox viruses that encode either the MART-1 or gp100 gene have also begun. Adjuvant administration of IL-2 is also being investigated based on murine studies.

To date, Dr. Rosenberg said, peptide immunization has been extremely effective in raising patients’ T-cell counts, but therapeutic response has been sporadic, as expected in phase I studies.

MART-1 showed immunogenic response in 100% of patients, he said, and one patient, treated with adenovirus altered to encode for MART-1, had a complete remission.