Moderna CEO Bancel served investors lemonade before the market opened on Monday at the interim phase 1 SARS-CoV-2 vaccine data conference. As of Tuesday’s market close the stock has nearly filled in Monday’s gap up. Bancel’s vague responses to very appropriate questions from analysts about antibody levels measured in the 1st 4 volunteers (of 45) vaccinated with mRNA-1273 (messenger RNA encoding the virus’s spike protein) strengthens my suspicion that this COVID-19 vaccine could be a lemon.
In the current era ushered in by cancer vaccine developers, computer modeling can predict precisely which small peptides (chains of amino acids) from a mutated - or foreign - protein can fit into a person's own chaperone molecule called MHC for presentation to a T lymphocyte which can then amplify the immune response. mRNA-1273 was not designed this way. mRNA-1273 encodes the virus’s large spike protein which is too large to be MHC-compatible without being processed in vivo. This will be left to chance in the hope that it will trigger an immune response. The potency of immune response is being measured by antibody levels (or titers, which means they get serially diluted until there is no longer a reaction in a test tube or petri dish). That has been standard procedure for years.
That doesn’t mean that currently available vaccines don’t recruit T cells to the fight against viruses, or that Moderna’s won’t. But it does mean that because mRNA-1273 isn’t designed to be a T cell (either the CD4+ helper type or the CD8+ cytotoxic type) antigen, it might not generate a cellular response. We won’t know because there is no provision to directly assay T cell response in vaccinated volunteers according to the protocol (NCT04283461) submitted with the IND application.
So while most vaccines are approved on the basis of antibody responses, some are more effective than others, and some flat out fail to generate a strong antibody response. That is why CEO Bancel’s comments carried so much weight - and weighed on that initial Monday morning surge in stock price.
Here are some important questions asked by analysts, and my excerpts from his fairly lengthy responses:
- “…could you provide us maybe a little bit more specificity around the specific titers…?” Excerpt of response: “…I don’t think it’d be appropriate to just give a number out there because what’s relevant is not the number…”
- “...how do you get confidence here as to what is protective for the binding and neutralizing antibody levels seen in any convalescent serum” [this refers to antibody titers in the group of persons recovering from COVID-19]? No numbers were provided in the CEO’s response. Some background here. Antibodies harvested from convalescents to treat other COVID-19 patients in China were at a levels of >1:640 (Duan, PNAS 2020). The FDA recommends titers of neutralizing antibody of at least 1:160 in donors whose plasma will be used for passive immunization to prevent infection for which that therapy is approved. So standards exist and it would have been helpful for Bancel to at least provide titers found the convalescent cases to give meaning to the phrase describing vaccinees' titers: "at or above levels seen in convalescent sera."
- Another analyst came back to the important point of antibody titers: “In the Phase 1 study for CMV, neutralizing antibody titers were around 10 to 40 times that of sera positive patients. And for the COVID-19 Phase 1, it seems to be much lower.” Again the response was evasive.
High titers of neutralizing antibodies have been found in the majority of patients recovering from COVID-19 but at various levels which were very low in 30% and in some undetectable (Wu, medRxiv 2020). So the presence of antibodies should not be assumed for the convalescent cases in Moderna's phase 1 safety trial in which titers are a secondary outcome measure.
Only one question about a cellular (i.e. T cell) response to the vaccine was asked at the conference. Here is part of the CEO's answer: "I personally find in the clinical trials that these are complicated finicky assays that really for an antigen like this and a disease like this, I don’t expect them to add that much of a significant information." "Finicky" means they require special expertise.
Moderna was unable to detect a cell-mediated response in a phase 1 trial (Feldman, 2019) its LNP–formulated mRNA vaccines against 2 strains of Avian influenza due either to lack of expertise or lack of T cell immunogenicity.
T lymphocytes are critically important actors in the immune response to virus infection. Antibodies can neutralize respiratory virus on the surface of cells that, e.g., line the human airway, but a strong antibody response requires help from so-called helper T cells (identified by the surface marker CD4). Other T cells (marked with the CD8 molecule) identify and kill tissue cells into which the virus has entered and set up a self-replication factory.
Given their importance relative to antibodies, it is surprising that T cells are hardly mentioned in all the public discussion about potential antibody therapy and preventive vaccination for coronavirus. But T cells are part of current preclinical vaccine research.
Grifoni et al. of La Jolla, CA (Cell 2020) assayed T cell response to SARS-CoV-2 virus antigens in 10 confirmed (by both PCR and antibody testing), convalescent cases of COVID-19. CD4+ T cells ex vivo responded with IFNγ (interferon gamma) production to spike protein and numerous other SARS-CoV-2 antigens in all 10. CD8+ T cells responded to virus antigens in 7/10 though the response was somewhat less vigorous than the CD4+ response (the ubiquitous CMV was used as a control virus for those comparisons). Spike protein antibody levels correlated with the individuals' spike-specific CD4+ T cell responses which is consistent with the role of CD4 T cells as 'helpers' of antibody secretion by B cells.
The La Jolla team found that the array of viral antigens responsible for the CD4+ T cell response was diverse - the spike protein, the usual target for coronavirus vaccine trials, accounted for only 27% of the total response. Spike protein antigens were responsible for only 26% of the CD8+ T cell response. Identifying antigens that would be recognized by helper CD4+ and cytotoxic CD8+ lymphocytes was a monumental effort facilitated by computer modeling. That modeling yielded 253 spike peptide and 221 non-spike peptide CD4-specific candidates, and 628 CD8-specific candidates. Vaccine candidate peptides were specific to either CD4 or CD8 T cells, and not necessarily both.
To my knowledge, neither the La Jolla team or any other has ever shown that virus-activated T cells actually do eliminate coronavirus-infected cells in vivo. The implication that they do comes from studies of patients infected with other respiratory viruses as reviewed here.
A better vaccine design. The La Jolla research shows that natural coronavirus infection elicits potent helper and cytotoxic T cell responses. Their results also suggest that a vaccine should include multiple antigenic targets not just the coronavirus spike protein. The ideal vaccine should include multiple, immunogenic (by ex vivo assay) CD4 and CD8+ T cell epitopes (portions of antigens) if they are to mimic natural immunity. The ideal vaccine should also produce immunity that is durable i.e. years not months - there again memory T cells may have an advantage over antibody-equipped memory B cells (see Channappanavar, Immunol Res 2014).
Inovio has an abiding interest in T cell vaccines because its main focus has been on curing gynecological cancers by vaccinating against HPV. Inovio is in phase 1 with its SARS-CoV-2 DNA vaccine called INO-4800. DNA delivery is via electroporation rather than by lipid nanoparticles (Moderna's). The DNA encodes the spike protein. Antigen-specific interferon-gamma (IFN-γ) cellular immune response and antibody titers are both primary outcome measures.
Inovio's GLS-5300 was the first MERS coronavirus vaccine to advance into human trials, and was found to be safe. "The vaccine-induced antibody and cellular immune responses were similar to those in the convalescent phase from samples in patients who recovered from natural MERS coronavirus infection...and poly-functional CD8+ T-cell responses that have been shown to correlate with less severe disease and lower MERS coronavirus shedding" (Modjarrad, 2019).
Inovio CEO Joseph Kim told Jon Cohen of Science that "a response by T cells -which clear infected cells - proved a better correlate of immunity [than antibodies] in monkey studies of the company's MERS vaccine."
So there is interest and expertise in designing coronavirus vaccines that leverage T cells in the amplification of the immune response. There must be quite a few. I haven't studied all of the COVID-19 vaccine developers. A figure from an article in Nature on 28 April portrays at least some of them in an array - enough to discuss in another article.
The chance that it will be Moderna that comes up with the most potent COVID-19 vaccine seems remote to me.
Disclosure: I/we have no positions in any stocks mentioned, but may initiate a short position in MRNA over the next 72 hours. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.