A precision medicine approach to developing cancer vaccines

Modern anticancer medicines represent a significant improvement on one-size-fits-all chemotherapy, reports Dr Sarah Houlton

Although non-specific cytotoxic drugs still have their place in the treatment of cancers, the ability to tailor therapy to the precise genetic make-up of a malignancy has greatly improved the prospects of many patients. Precision medicine strategies range from drugs that target specific antigens that are overexpressed on the surface of cancer cells to vaccination programmes that prime the body’s own immune system to fight back.

The number of cancer vaccine products already on the market is limited. The first to be approved in the US, back in 2010, was sipleucel-T (Provenge) from Dendreon, now owned by Valeant. The vaccine is based on autologous dendritic cells, harvested from the patient’s blood via leukapheresis. These are then cultured with the protein PAP-GM-CSF, of which the granulocyte-macrophage colony stimulating factor component stimulates the immune system, and enhances the presentation of the antigens.

The only other cancer vaccine to be given FDA approval, talimogene laherparepvec (Imlygic), originally developed by BioVex and acquired by Amgen, was ratified in late 2015. This is an oncolytic viral therapy, and is a genetically engineered form of the cold sore virus, herpes simplex-1. It has been attenuated, its selectivity to cancer cells was increased, and it was also engineered to secrete GM-CSF. It is licensed for the treatment of melanoma, and is directly injected into the tumours; it also has systemic effects.

Very many more cancer vaccines are in the early stages of development. Indeed, at this June’s ASCO meeting in Chicago, USA, more than 50 abstracts mentioned the term ‘vaccine.’ And a number of those vaccines are looking promising. For example, scientists at the University of Washington, with the backing of the US National Cancer Institute, are investigating a DNA plasmid vaccine that targets insulin-like growth factor binding protein-2, or IGFBP-2. This self-antigen plays an important role in regulating the invasiveness and development of metastases in ovarian cancer.

The idea is that by immunising against these self-antigens, regulatory responses might be induced that inhibit the development of desirable type I antitumour immunity. If epitopes that bias towards a regulatory phenotype are removed, vaccine efficacy might be improved.

A single arm, non-randomised Phase I study was done in 25 patients with advanced or recurrent ovarian cancer to evaluate its safety and immunogenicity.1 Subjects were given three monthly doses of an IGFBP-2 DNA vaccine via intradermal injection, and IGFBP-2-specific T-cell precursor frequencies were significantly elevated compared with baseline levels at 4 and 6 months, whereas T-regulatory cells were no higher than the numbers measured in a control group. The overall survival rate after 2 years was 82%, and the median overall survival rate had yet to be reached during a median follow-up period of 23.5 months. Fatigue and injection site reactions were the most common adverse events. They concluded that it is well tolerated and generates significant Type I immunity.

mRNA and glycoproteins

In projects licensed from German biotech CureVac, Boehringer Ingelheim is looking at the potential of mRNA components as cancer vaccines. The furthest advanced, CV9202 for non-small cell lung cancer, targets six different tumour associated antigens: 5T4, Mage C1, Mage C2, NY-ESO-1, survivin and MUC1. The hope is that this will stimulate the immune system to destroy lung cancer cells. It is being investigated in combination with, variously, afatinib, a checkpoint inhibitor and radiotherapy.

The interim analysis of a Phase Ib trial in conjunction with local radiotherapy looks promising.2 In all, 26 patients with stage IV NSCLC were enrolled in three cohorts — squamous and non-squamous both with and without activating EGFR mutations. They were given two vaccinations before local radiotherapy to a single tumour lesion in four consecutive daily fractions of 5GY. The vaccination was continued until subsequent anticancer therapy was initiated, and maintenance doses of pemetrexed or EGFR inhibitors where appropriate.

The median number of vaccinations was seven; 15 patients received pemetrexed and two EGFR inhibitors. No vaccine-related serious adverse events were reported. There was one confirmed partial response in a patient on pemetrexed and, in all, 13 experienced a stable disease condition. Tumour shrinkage of at least 15% at sites outside the area irradiated occurred in seven of 25 assessable patients.

OBI-821/OBI-822, being developed by Taiwanese company OBI Pharma under licence from the Memorial Sloan-Kettering Cancer Center in the US, is a synthetic glycoprotein. The two-component vaccine includes a tumour-associated carbohydrate antigen, Globo H, which is covalently bound to a carrier protein, keyhole limpet haemocyanin, to enhance its immunogenicity (the OBI-822 part), and OBI-821, a saponin-based adjuvant. Globo H is overexpressed on the surface of various epithelial cancers, such as breast, colon, gastric, lung, ovarian, pancreatic and prostate.

A Phase II/III trial has been done to evaluate its clinical benefit and immunogenicity in a cohort of patients with late-stage metastatic breast cancer.3 Subjects were given nine injections of the vaccine or a placebo during a period of 41 weeks or until disease progression occurred, in combination with low-dose cyclophosphamide; hormone therapy was also permitted. The vaccine did not meet its primary endpoint of progression-free survival, but it did elicit a significant improvement in progression-free survival in the 50% or so of patients who demonstrated an immune response to the vaccine therapy, compared with the placebo group. Overall survival data is not yet available, and a Phase III trial is planned.

The T-cell approach

Basel-based Vaxxim is developing an oral T-cell vaccine, VXM01, designed to target the VEGF-receptor 2 antigen on tumour vasculature and immunosuppressive cells. Based on an approved live attenuated salmonella vaccine, the bacteria are modified to create a eukaryotic expression plasmid encoding the genetic information of the target antigen. This causes patients to develop specific cytotoxic T-cells against these targets after dosing.

The first clinical trial was in pancreatic cancer. An initial trial, in which four priming vaccinations were given within a week, showed that VEGFR2-specific T-cell responses gradually declined after the peak response was achieved at day 21, so an extension trial was done to see if boost vaccinations might be successful.4

A group of 26 patients with advanced pancreatic cancer were given the priming regimen, and then 6-monthly boost vaccinations starting on day 38, or a placebo in a randomised Phase I trial. Of the 18 vaccinated patients, 12 showed considerable, or even high, increases in VEGFR2-specific T-cell response compared with baseline. Across the entire vaccinated group, levels peaked after 3 months, with an average four-fold increase above baseline, and they were still above baseline after 6 months. Median overall survival was 9.3 months for the vaccinated group, and 8.4 months for those given the placebo.

The lead product from Nova Scotia-based Immunovaccine, DPX-Survivac, aims to train the immune system to generate CD8 T-cells against the tumour-associated antigen survivin, which acts as an apoptosis inhibitor and is associated with a large number of both solid tumours and haematological malignancies. Increased expression has good correlation with progression and the development of drug resistance. The vaccine comprises a mix of surviving HLA class I peptides that have been formulated to evoke a cytotoxic T-cell response against survivin.

A Phase I study showed its immunogenicity in combination with metronomic low-dose oral cyclophosphamide.5 In a Phase Ib trial to determine optimal dose and treatment schedule for this combination, 36 patients with ovarian, fallopian tube or peritoneal cancer who experienced a complete or partial response after first or second line chemotherapy and who had minimal disease were given various vaccination schedules, as well as boosting without the oil component of DPX-Survivac.6

The clinicians report that sustained immune responses of varying magnitude and duration were achieved after several modifications of the dose and schedule were made. Two randomised Phase II trials are planned, where it will be given in combination with an immune modulator in ovarian cancer and glioblastoma.

Peptide potential

Japanese company OncoTherapy Science has developed a peptide vaccine cocktail, OCV-C01, which contains epitope peptides derived from VEGFR1, VEGFR2 and KIF20A. In a Phase II trial, it was given in combination with gemcitabine, with the aim of estimating disease-free survival and exploring surrogate biomarkers.7 In all, 30 patients with pancreatic ductal carcinoma who had undergone pancreatectomy were given weekly doses of the vaccine for 48 weeks, plus gemcitabine on days 1, 8 and 15 of a 28-day cycle.

Median disease-free survival was 16 months, and the disease-free survival rate at 18 months was 35%. Subgroup analysis indicated significant differences in disease-free survival for those with positive CTL responses specific for KIF20A peptides and without them, and also between those positive for KIF20A expression and those who were not. All four of those patients with this expression had no recurrence of their cancer, and their disease free survival was significantly longer than those without.

WT1, or Wilms’ tumour gene, is another target proving popular in vaccine studies. A Phase II study at the department of cancer immunotherapy at Osaka University in Japan looked at the potential of a peptide WT1 vaccine they had developed in advanced pancreatic ductal adenocarcinoma.8 The vaccine comprises an HLA-restricted nine-mer WT1 peptide, plus Montanide ISA51 as adjuvant, and is administered intradermally.

Many cancer vaccines are in the early stages of development; at this June’s ASCO meeting in Chicago, USA, for example, more than 50 abstracts mentioned the term ‘vaccine’

The 91 patients, who had locally advanced or metastatic disease, without prior treatment or recurrence after surgery, were randomised to receive the vaccine in combination with gemcitabine, or gemcitabine alone. Those who progressed on gemcitabine alone were permitted to receive the vaccine. The vaccine was given on days 1 and 15, and gemcitabine on days 1, 8 and 15, of a 28-day cycle until disease progression occurred. Those given the vaccine had a significantly improved one-year overall survival rate, and a prolonged progression-free survival, compared with those given gemcitabine alone.

The clinicians described the results as ‘remarkable in patients with metastatic pancreatic ductal adenocarcinoma.’ Median progression-free survival was 195 days in those vaccinated patients found to have delayed-typed hypersensitivity to WT1 peptide, compared with 102 days for those who did not have this hypersensitivity, and 100 for the placebo group.

Swiss-headquartered Sellas Life Sciences also has a WT1 vaccine. Galinpepimut-S is a multivalent peptide vaccine, using both native and synthetic WT1 peptide sequences, that was designed to stimulate both CD4 and CD8 T-cell responses. Promising results have been reported in malignant pleural mesothelioma.9 Within 12 weeks of completing multimodality therapy, 40 patients were randomised to receive Montanide and GM-CSF, either with or without the vaccine, with six vaccinations being given at 2-weekly intervals. The median progression-free survival from randomisation was 11.4 months in the vaccinated arm, and 5.7 in the non-vaccinated group.

Further work

Good results are also being seen in patients who had achieved a complete response in acute myeloid leukaemia.10 Again, the dosing schedule was six 2-weekly vaccinations with the adjuvants Montanide and GM-CSF; six further monthly doses could be given if they remained in complete response. Of the 22 patients in the study, 14 completed at least six vaccinations, and 10 received all 12. At the point of analysis, 14 had relapsed, and 10 died from disease progression. The median leukaemia-free survival was almost 2 years, and the overall survival nearly 4 years. Survival was shorter for those who did not have an immunological response to the vaccine. A pivotal randomised study is planned.

A 5-year follow-up on metastatic melanoma patients who had been treated with Caladrius Biosciences’ vaccine eltrapuldencel-T was promising.11 The vaccine comprises autologous dendritic cells loaded with irradiated autologous cancer-initiating cells, allowing the entire spectrum of antigens from the patients own tumour to be presented.

Two previously reported Phase II trials showed a 72% 2-year overall survival, and the median overall survival was more than twice as long as was seen with a vaccine consisting of irradiated autologous tumour cells. During the period 2001–2011, 72 patients with stage IV or recurrent stage III melanoma were treated with the vaccine in these trials. All 33 surviving patients were followed for 5 years, with a median overall survival of almost 3 years. The 5-year overall survival rate was 46%, and 19% was progression-free.

A rather different strategy was taken by a team at the City of Hope Medical Center in Duarte (California, USA) – a modified vaccinia Ankara virus containing the wildtype p53 transgene p53MVA. The aim is to stimulate immunity to overexpressed p53 in tumour cells. Initial results from a Phase I study in ovarian cancer are promising; an earlier Phase I trial in gastrointestinal malignancies were similarly promising, with a well tolerated dose having been established.

In the light of good results with alternative cancer vaccines administered in combination with gemcitabine, they carried out a de-escalation Phase I trial of p53MVA plus gemcitabine in patients with ovarian cancer.12 At the time of reporting, seven patients with a platinum-free interval less than 12 months and who had previously been given no more than two prior chemotherapy lines for recurrent disease, and histology showing p53 involvement. They were given gemcitabine on days 1 and 8 of three-week cycles, plus the vaccine on day 15 of the first three cycles.

No dose-limiting toxicities were seen, and therefore the starting dose was established as appropriate for subsequent Phase II trials, and additional patients are being recruited at this dose. The clinicians believe this is the first trial to have been done in ovarian cancer that combines a viral-based cancer vaccine and gemcitabine in ovarian cancer.


1. J.B. Liao, et al., J. Clin. Oncol. 34(Suppl.), Abst. 5542 (2016).

2. M. Sebastian, et al., J. Clin. Oncol. 34(Suppl.), Abst. e20627 (2016).

3. C-S. Huang, et al., J. Clin. Oncol. 34(Suppl.), Abst. 1003 (2016).

4. F.H. Schmitz-Winnenthal, et al., J. Clin. Oncol. 34(Suppl.), Abst. 3091 (2016).

5. N.L. Berinstein, et al., Oncoimmunology 4, e1026529 (2015).

6. J.A. Villella, et al., J. Clin. Oncol. 34(Suppl.), Abst. e14577 (2016).

7. H. Yamaue, et al., J. Clin. Oncol. 34(Suppl.), Abst. e14587 (2016).

8. S. Nishida, et al., J. Clin. Oncol. 34(Suppl.), Abst. 3085 (2016).

9. M.G. Zauderer, et al., J. Clin. Oncol. 34(Suppl.), Abst. 8519 (2016).

10. P.G. Maslak, et al., J. Clin. Oncol. 34(Suppl.), Abst. 7005 (2016).

11. R.O. Dillman, et al., J. Clin. Oncol. 34(Suppl.), Abst. 3090 (2016).

12. M.C. Cristea, et al., J. Clin. Oncol. 34(Suppl.), Abst. e17040 (2016).