Phase I clinical trial of a recombinant canarypoxvirus (ALVAC) vaccine expressing human carcinoembryonic antigen and the B7.1 co-stimulatory molecule

Authors: Horig H, et al.
Source:Cancer Immunol Immunother 49, 2000, pg 504

Background

• Vaccination strategies for cancer treatment has been an idea for nearly a century, but without much success
• New understandings of immune function and immune responses have led to a resurgence of cancer vaccines
• The new insights are that T-cells specific for tumor antigens can be generated if presented correctly. Further, antigen presenting cells (APCs) such as dendritic cells, play a central role in this presentation and stimulation of T-cells. A subsequent immune response requires the delivery of a costimulatory signal, also emanating from the APC.
• This study uses carcinoembryonic antigen (CEA) as the target antigen, which is overexpressed in many adenocarcinomas, with primary sites including the pancreas, colon, rectum, stomach, lung, and breast.
• This was the first study to also employ the costimulatory molecule B7.1
• Both were delivered in the vaccine by the ALVAC virus, a nonreplicating virus that elicits a strong T-cell reaction
• As this is a phase I trial, the objectives were to determine the dose and toxicity of increasing doses of the ALVAC-CEA B7.1 vaccine

Materials and Methods

• Patients with measurable and unresectable adenocarcinoma expressing CEA were eligible for the study
• This was a dose escalation study, with 6 patients receiving 4.5 x 10 6 plaque forming units of viral vaccine, 6 patients receiving 4.5 x 10 7 plaque forming units, and 6 patients receiving 4.5 x 10 8 plaque forming units
• Vaccination was given intramuscularly every four weeks for three total doses
• If there was no evidence of disease progression, another series of vaccinations was offered
• Evaluation of treatment response was determined by measuring the volume of radiologically defined tumor lesions. Patients without change in size or <25% increase in their lesion size were recorded as having stable disease
• Evidence of grade 3 or greater nonhematologic toxicity was used to stop any further treatment
• The occurrence of two grade 3 toxicities at a given dose level defined the dose-limiting toxicity
• Antibody responses against ALVAC were tested using enzyme linked immunosorbent assay (ELISA)
• The intensity of immune response, by measuring the frequency of CEA-specific T-cells, was done using enzyme linked immnoassay spot tests (ELISPOT)
• 18 patients were enrolled at the above mentioned dose levels
• All patients except one had received extensive prior treatment, including surgery, chemotherapy, and radiation therapy.

Results

• Two patients could not be fully evaluated, as one patient died two weeks after starting the trial and one patient withdrew from the study
• No dose limiting toxicities were seen at any dose level. In fact, clinical toxicity was minimal except for mild pain at the site of injection
• Seven patients had a mild increase in alkaline phosphatase and hepatic transaminase levels
• Thrombocytopenia developed in 3 patients, yet none were severe
• One patient developed an exacerbation of pre-existing self-limiting psoriasis after the first dose of vaccine. No other signs of autoimmunity were noted, including no increases in ANA titers throughout treatment
• Antibody titers to ALVAC were increased by three vaccinations in most patients. The greatest increase was observed in sera from patients given the highest dose of vaccine
• Because of problems inherent with the ELISPOT test, only 12 patients were evaluable for the generation of CEA specific T-cells. Four patients showed an increase in T cells. Three of these patients were the only patients in the study that experienced stable disease
• These three patients maintained their T-cell responses when a second cycle of vaccine was given

Authors' Conclusions

• This study represents the first use of B7.1(the co-stimulatory molecule) co-expressed with a tumor antigen (CEA) in human patients
• The vaccine was safe, with few side effects noted and no maximally tolerated dose level was reached
• The vaccine elicited measurable immune (T-cell) responses in four patients. Three of these same patients experienced stable disease
• Additional studies with larger numbers of patients are warranted

Scientific Implications

This study does represent the first use of a novel approach to cancer vaccination. The administration of a co-stimulatory molecule appears to be key in animal and laboratory studies. Therefore, this study tested a vaccination with the costimulatory molecule B7.1 with an anti-CEA vaccine. As this was a phase I trial, most importantly, there were no major toxicities and no evidence of autoimmunity. The concern over the generation of an autoimmune response was warranted, as CEA is expressed on much of the intestinal epithelium. The efficacy of the vaccine, however, is in question. Granted, this was a phase I trial, not designed to investigate efficacy per se, but the authors indicate that the responses seen were encouraging. This is true from the standpoint that all of these patients had been treated many times before, likely with resultant dulled immune systems. They also had metastatic disease, likely unresponsive to conventional intervention. From that respect, the response in three patients is excellent. However, the definition they used for stable disease allowed the tumor to progress up to 25%. In addition, a T-cell immune response was seen in only 4 patients. Though this is an indirect measure of the efficacy of the vaccine, it is still an indication that the vaccine may require further refinement. Regardless, in the complex field of immunology and cancer vaccines, the fact that this study used a tumor specific antigen laden vaccine with the novel addition of the costimulatory molecule makes it an important step in the development of an effective cancer vaccine.