Originally published July 9.
A European-led consortium of researchers and biotech firms are working together to develop and manufacture individualized vaccines based on the genomic, proteomic, and peptidomic characteristics of a particular glioblastoma patient's tumor and immune system.
Launched in late 2012, the Glioma Actively Personalized Vaccine Consortium, or GAPVAC, has received a €6 million ($7.7 million) European Union Framework 7 grant to conduct an early-phase trial in which 30 newly diagnosed glioblastoma patients are treated with cancer therapeutic vaccines developed based on biomarkers underlying their disease. The Phase I trial will begin recruiting patients next year. As the name of the consortium suggests, researchers involved with the project are hoping to demonstrate that cancer treatments can be "actively" personalized for glioblastoma patients, instead of being passively targeted for their illness.
"When we talk about personalized medicine today, you have a certain arsenal of therapies on the shelf and you try to find the right patient who fits into that arsenal. And if there is nothing in that arsenal that fits the patient then the patient doesn't really have any therapeutic option," Harpreet Singh, CSO of Immatics Biotechnologies, told PGx Reporter. Through GAPVAC, "what we try to do is to look into the cancer of the individual patient and actually create a therapeutic vaccine that's directed at that specific individual disease."
GAPVAC involves 14 organizations, including Immatics, a spin-off from the University of Tübingen, and BioNTech, a spin-off of Johannes Gutenberg-University Mainz. These two German biotech firms are coordinating the project, and will likely have rights to certain IP resulting from the four-year effort.
With its antigen discovery method, called XPRESIDENT, Immatics will generate a warehouse of tumor-associated peptides. The XPRESIDENT discovery engine came out of the Eberhard Karls University Tübingen, where immunologist Hans-Georg Rammensee and his research team established the so-called "Tübingen approach" of identifying, selecting, and validating HLA-associated peptides in tumor-associated antigens.
The approach involves identifying naturally present HLA-associated peptides from primary tumor cells; selecting overexpressed peptides in tumors based on gene expression analysis and bioinformatics analysis; and then picking validated peptides by monitoring the patient's T-cell responses in vivo. Using this method, Rammensee's research showed that the tumor tissue from different patients looked very different in terms of the antigens, or the immune structures, and that these differences could be therapeutically targeted.
Meanwhile, researchers led by Christoph Huber, chair of the hematology/oncology department at Johannes Gutenberg-University in Mainz, discovered that cancer patients' naturally existing immune responses were directed against mutated antigens and that these mutated antigens were unique to a particular patient's tumor.
"If you merge these two observations" from Rammensee and Huber's research, "it makes a lot of sense actually, thinking about how we could further drive efficacy of an immunotherapy approach," Singh said. "One way to go, we think, is to actually adapt the product to the patient, and not the other way around like it's done today."
In GAPVAC, researchers will be able to create personalized immunotherapeutics directed at tumor-associated antigens and tumor-specific antigens. Tumor-associated antigens are present mainly on malignant cells but may also be expressed on some normal cells. Tumor-specific antigens are present only on tumor cells
After analyzing the overexpressed tumor-associated antigens and immunogenicity in a particular patient, researchers in GAPVAC can give that study participant a personalized vaccine developed with pre-manufactured, off-the-shelf peptides from Immatics' peptide warehouse.
BioNTech, meanwhile, has preclinically validated a process of advancing individualized vaccines, which uses next-generation sequencing approaches to identify the genomic mutations in a patient's cancer that give rise to tumor-specific antigens. After identifying these mutations, the company rapidly synthesizes the vaccine directed at these tumor-specific antigens, administers it to the patient, and monitors the patient's immune response to the vaccine.
In GAPVAC, BioNTech will identify genomic mutations in patients using next-generation sequencing and Immatics will confirm that these mutations have led to mutated peptides in the patient. Using BioNTech's vaccine development approach, GAPVAC researchers can administer study participants a treatment that is essentially made to order for that person, specifically targeting the uniquely mutated antigens in his or her tumor.
BCN Peptides in Spain will manufacture the warehouse peptides, while the final formulations of the peptides will be manufactured at the University of Tübingen's Department of Immunology. Wolfgang Wick at the University of Heidelberg and Pierre-Yves Dietrich at the University of Geneva will lead the Phase I clinical trial. The non-profit, Association of Cancer Immunotherapy, will lead the biomarker program for the trial, with Immatics identifying the marker signatures for each patient.
Patients for the trial, who will be treated with these actively personalized vaccines, will be recruited from academic institutions in Europe and the US, including Eberhard Karls University Tübingen, Germany; Beatson West of Scotland Cancer Centre; Universities Hospital Geneva, Switzerland; Universities Hospital Heidelberg, Germany; Herlev Hospital, Denmark; Leiden University Medical Centre, The Netherlands; University of Pittsburgh Cancer Institute; University Southampton, UK; Technion, Israel; and Vall d’Hebron University Hospital, Spain.
In Singh's view, GAPVAC's effort to personalize cancer immunotherapeutics was made possible by advances in diagnostics technologies. "Now is the right time," to actively individualize cancer vaccines, "because 10 years ago when [researchers] first had these ideas, we didn't really have the tools to look at this the way we can today," he noted. "We have technologies today that make [cancer] heterogeneity more accessible and more readable."
GAPVAC researchers will use mass spectrometry to assess the range of antigenic peptides in patients' cancer and use a combination of mass spec and next-generation sequencing to characterize genomic mutations leading to mutated antigen peptides in study participants' tumors. By comparing the genome sequence of a patient's cancer cells and healthy cells, GAPVAC researchers will "identify what's different in the cancer, the mutations [which] form proteins that are new and different … [which] develop into new and different peptides. These peptides can then be targeted by the immune system," Singh explained.
Previously, not having access to advanced diagnostic tools and the limited knowledge about targetable tumor antigens held back early attempts to personalized cancer immunotherapeutics, Singh observed. "I'm saying this a little crudely, but [previously drug developers] took a tumor, put it into a blender, and gave it back to the patient in the hopes that the parts that are in the tumor soup may activate the immune system," he said. "The issue with that approach is that among the millions of non-relevant proteins and peptides, a tiny fraction is really tumor associated … What we're able to do in the era of peptidomics, proteomics, and genomics is really identify the relevant target structure and then amplify that by making it synthetically available to the body."
Singh characterizes earlier cancer vaccines as being "passively" personalized. "These tumor lysates or orthologous vaccines are personalized for the patient but no one knows actually what's in there," he said. "What we're doing, we're calling actively personalized because it's really driven by the biomarkers we identify on the tumor tissue."
Still, GAPVAC is at an early, experimental stage, where researchers are hoping to demonstrate that this actively personalized vaccine approach is feasible in a clinical trial involving a limited number of patients. Real-world application to quickly develop vaccines for an individual patient may be costly and will require a level of coordination between stakeholders that are likely not used to working together in the commercial realm.
"Once this project is successfully completed and shows that it's feasible in principle to do such an approach, then we have to think more about how to bring that to a commercial platform," Singh said, predicting that real-world application will be possible by combining the two drug development concepts being used in GAPVAC – the off-the-shelf tumor-associated peptide warehouse and "on demand" vaccine synthesis.
Conceivably, in the commercial setting, after a patient is analyzed for biomarkers, drugmakers could put together a vaccine using pre-manufactured peptides to treat that patient. "This can be developed fairly quickly," Singh said. "And we believe this is something that is absolutely possible to commercially explore."
It will be more challenging, Singh acknowledged, to commercialize development of "on demand" personalized vaccines that target antigens based on tumor-specific mutations. "You cannot pre-manufacture that vaccine," he said. "It is manufactured specifically for one patient and there is a high probability that there won't be another patient that will use the same peptide."
This approach will require a company to be able to synthesize the vaccine within a few months, which BioNTech has simulated in pre-clinical models. At this time, however, Singh said it was too early to provide an estimate of what it might cost to develop such individualized vaccines and administer them to each of the study participants in the GAPVAC Phase I trial.
"In principle, when working with synthetic antigens, like peptides or RNA, there are lots of factors that are easily scalable," Singh said. "That's in contrast to orthologous vaccines or tumor lysates where you have to go through the same process every time. That kind of scalability can dramatically reduce your cost. So, we see further optimizing potential in the next years."
GAPVAC is currently in the preclinical phase, and study investigators are in discussions with regulatory authorities and are setting up the infrastructure for the trial, such as the peptide warehouse. According to the project website, one of GAPVAC's aims is to establish a regulatory framework for advancing actively personalized vaccines. The project is hoping to enroll the first patient in the second quarter of 2014.
GAPVAC has chosen to initiate development of actively personalized cancer vaccines in glioblastoma, specifically, due to the unmet medical need in this disease setting. After diagnosis, glioblastoma patients live for around a year with the help of radiation treatment and temozolomide chemotherapy, the standard of care.
At the American Society of Clinical Oncology's annual meeting in June, disappointing results from several studies investigating glioblastoma treatments highlighted the drug development challenges in this space. Roche/Genentech's Avastin (bevacizumab) failed to extend overall survival in newly diagnosed patients but improved progression-free survival. Biomarker analysis in a separate study funded by the NCI has shown some promise in being able to identify which glioblastoma patients might have a favorable response to an Avastin-containing regimen.
Also at ASCO, Merck released data from a failed Phase III trial investigating an integrin inhibitor called cilengitide in newly diagnosed glioblastoma patients whose tumors harbored MGMT gene promoter methylation. Despite a large biomarker testing program, Merck was unable to identify a subgroup that responded to the drug (PGx Reporter 6/5/2013).
However, cancer immunotherapeutics are a major focus and a big area of investment for drug developers. In that regard, glioblastoma may be a particularly promising area for personalized immunotherapy development, Singh believes. "When we looked at glioblastoma, we found a number of properties in this cancer that we believe make this a very good target for cancer immunotherapy," Singh said.
In a paper published in Brain, researchers led by Valerie Dutoit of the University of Geneva reported that after analysis of 32 HLA-A*02-positive glioblastoma samples by mass spectrometry, they identified more than 6,000 HLA-restricted peptides comprising over 3,000 different HLA-A*02-restricted sequences. By assessing the high overexpression of messenger RNA in glioblastoma samples, researchers identified 10 of the most immunogenic HLA-restricted peptides.
"Glioma tumors are rich in immunogenic antigen," said Singh, who was one of the authors on the Brain paper. "There are a lot of antigens and many of these antigens make the immune system respond well against these antigens."
Using some of these highly immunogenic peptides identified in the study, Immatics is developing the multi-peptide therapeutic vaccine, IMA950. The firm has launched two Phase I trials in the UK and in the US to test the immunogenicity of the vaccine. In these studies, investigators will gauge gloiblastoma patients' biomarkers and correlate them to their clinical and immunological responses to the vaccine.