With personalised cancer vaccines on the rise, is manufacturing scalable yet?

The COVID-19 pandemic reminded us how vaccines can stop disease in their tracks and challenged the notion that vaccines take many years to develop. These new vaccine modalities are also more readily altered to suit evolving pathogens. This improved speed and flexibility is now being taken to the extreme for personalised cancer neoantigen vaccines.

The Covid-19 pandemic showed the great promise of vaccine platforms that have a common structure but can be easily modified to expose the immune system to different antigens. For example, the Moderna and Pfizer mRNA vaccine and the Astra-Zeneca Adenovirus vaccine were quickly updated for new Covid-19 variants.

At the same time, the power of the immune system is increasingly being used to fight cancer. Great strides have been made with approaches such as checkpoint inhibitor drugs and CAR-T therapies. However, while these treatments give good results for some cancers, for more intractable cancers a method of making these tumour types more ‘visible’ to the immune system is needed. One hugely promising approach is personalised neoantigen vaccines.

Personalised neoantigen vaccines are an example of extreme customisation in vaccine platforms, to the level of individual patients. However, there is a risk that these therapies, even now progressing through clinical trials, may not reach the widest number of people, unless several screening, selection and manufacturing challenges are addressed.

In this TTP Market Insight, we look at the biological background behind neoantigen vaccines and note that new GMP-standard equipment and consumables, new approaches to quality control, and novel automation and liquid handling solutions will be required to usher in this future of personalised cancer vaccines.

Cancer cells are often genetically unstable and, through various mutational pathways, both point mutations and chromosome alterations are induced. Some but not all of these mutations can generate “neoantigens”, that is novel proteins not already present in the patient’s normal cells.

For this to happen the mutation must be within a coding part of the genome and must change the amino acid specified. The mutated protein must also then be expressed (translated from mRNA into protein) in the tumour cell. Finally, the ability of the neoantigen to stimulate an immune response depends on how the mutated protein is processed into smaller peptides and if these fragments are amenable to MHC antigen presentation on the cell surface.

Owing to their “non-self” origin, neoantigens have the potential to be highly immunogenic and thus clear targets for immunotherapy. Moreover, immunotherapies targeting neoantigens are less likely to induce autoimmunity and generate unwanted side effects. Add to this the fact that vaccines can produce a long lasting immune ‘memory’, and the result should be a vaccine that enables treatment of the initial disease and then prevention of a re-occurrence.

Whilst individuals suffering from the same cancer type might exhibit some similar patterns of genetic alterations in the affected cells, studies show that the vast majority of neoantigens are patient-specific, therefore necessitating a personalised approach. The rapid advancement of next generation sequencing technologies allows quick and relatively inexpensive identification of tumour associated mutations.

However, identification of tumour associated mutations does not guarantee effective treatment since cancerous tumours tend to be heterogeneous in nature (non-clonal) and not all of the tumour tissue may have the same mutations causing neoantigens. To effectively treat the disease, a neoantigen must be chosen that is present in all cancer cells otherwise the patient may relapse. Therefore, it can be seen that the pool of mutations that might lead to a neoantigen generating a strong immune response is only a small subset of the total number.

One of the major challenges, therefore, of personalised neoantigen therapy is selection of the most effective neoantigen candidates ahead of vaccine manufacture. For large numbers of patients to be treated effectively, this would require a high throughput platform capable of integrating all the relevant information. This platform would need sophisticated prediction methods to whittle down all the genetic information to just a few candidates. This screening and prediction methodology would form an integral part of the treatment and would need to be clearly defined to regulatory bodies.

To date, various vaccine modalities have been trialled in the clinic, to translate a neoantigen ‘fingerprint’ into a therapy, each stimulating the immune system in a distinct manner.

Antigen presenting dendritic cells have featured in a number of vaccine trials. This process typically involves harvesting of dendritic cells from a patient and exposing them to the antigen in an ‘ex vivo’ process. The primed dendritic cells can then be returned to the patient where they stimulate an immune response.

Another commonly used modality in this type of vaccine is free peptide that corresponds to a processed and presented antigen. As naked peptides are not in themselves particularly immunogenic, this approach generally also requires the use of an adjuvant.

Before the Covid-19 pandemic, mRNA vaccines had also been trialled for cancer vaccines. Thanks to the huge success of the Moderna and Pfizer-BioNTech vaccines, the potential of mRNA vaccines is even more clear, and this modality may become a front runner in personalised neoantigen vaccines.

Regardless of which vaccine modality is employed, the personalised nature of neoantigen therapies throws up a set of similar manufacturing challenges.

To be able to benefit the maximum number of people, hundreds or thousands of different vaccines must be produced in parallel. The manufacturing process required for this is hugely different to that of more standard vaccines where large bioreactors and downstream processing methods are used.

At the moment, the commercially available equipment and consumables suitable for small batch processes are typically only used for research and development purposes. Therefore, these items are generally not suitable for pharmaceutical manufacturing use. For personalised neoantigen vaccines, a whole new generation of GMP grade, highly parallel small processes need to be developed, placing new demands on equipment and consumables.

We might be able to learn from the challenges presented by the small batch sizes and personalised nature of autologous cell therapies.

TTP spin-out company “Cellular Origins” is building a scalable manufacturing solution for cell and gene therapy. The ConstellationTM platform leverages robotic automation to provide the capability to manufacture these otherwise labour-intensive therapies at factory scale, parallelising many of the unit operations while maintaining closed, sterile operation.

In leveraging our understanding of the equipment design requirements for GMP deployment, TTP worked with CellFiber to transform their R&D prototype into a patient-scale cell therapy manufacturing device, designed to provide high-throughput, low-shear cell processing in a GMP environment. Some of the design requirements for personalised neoantigen vaccines are likely to be similar.

Additionally, the landscape of QC testing and quality assurance will have to adapt to accommodate this new manufacturing paradigm. Traditional release testing of pharmaceutical batches often requires quite large samples of product. This is not a problem when the overall batch size is large and the samples represent a small percentage of the production volume. However, when your batch is for an individual patient, methods of QC testing will have to adapt to match the much smaller sample volume that may be available. Some QC tests take a long time to initiate and may take many weeks to complete. This is just not feasible when a very sick patient is waiting for their treatment.

In our next article, we will consider in more detail how know-how from diagnostics – an industry to which TTP has long been a partner – can be helpful in developing distributed analytical solutions to deliver on the promise of personalised therapies for patients.

Clinical trials of personalised neoantigen vaccines are typically conducted in late-stage patients who have already been treated with other therapy types. However, if neoantigen vaccines continue to show promise, it may be more impactful to treat earlier stage patients. This could be beneficial because earlier stage patients may not have the pronounced tumour heterogeneity of late-stage disease and therefore a neoantigen vaccine might be better able to tackle all abnormal cells.

Additionally, earlier stage patients have more time available for therapy developers to conduct neoantigen identification and vaccines production. It is also likely that neoantigen vaccines would be used alongside other treatments such as checkpoint inhibitors so they can act in a synergistic way. The earlier stage and combined use of neoantigen vaccines would make the pool of potential recipients much larger and therefore the manufacturing capacity challenge much greater.

Personalised neoantigen vaccines are showing great promise. But to be of benefit the maximum number of people, neoantigen identification pipelines, vaccine manufacturing processes and quality control need to be streamlined. The R&D effort required to support this needs to get underway well before clinical trials are complete. Though the challenges are sizeable, in an ideal world, neoantigen therapy could start with a biopsy and end with patient vaccination for all that need it.