Viralytics — Looking towards pivotal Cavatak combo studies

Viralytics is developing the oncolytic virus, Cavatak, for use in immunotherapy treatments for a range of cancers. Cavatak is a proprietary formulation of a common cold virus, Coxsackievirus A21 (CVA21), in clinical development for late-stage melanoma, lung and bladder cancer, with potential applications in a range of other cancers. When injected into melanoma lesions, Cavatak achieved a 28% response rate as a single agent, and preliminary response rates of 60% and 57% when combined with ICI drugs Keytruda and Yervoy, respectively. Adverse events rates have been low in all Cavatak trials so far, even when combined with therapies usually associated with high adverse event rates. Initial results from the Keynote 200 Phase I/II study (in collaboration with Merck) produced encouraging signs that Cavatak will also be effective when administered intravenously (iv); iv Cavatak is being trialled in combination with Merck’s Keytruda in lung and bladder cancer.

We value Viralytics at A$469m, or A$1.95 per share (undiluted), using a risk-adjusted net present value method to discount future cash flows through to 2044 in metastatic melanoma, bladder cancer, non-small cell lung cancer, head and neck cancer and metastatic colorectal cancer (mCRC). We apply a 40% probability of success to intralesional injection of Cavatak in melanoma to reflect positive clinical data and the clear market opportunity for Cavatak plus Yervoy in patients who have failed PD1/L1 ICI therapy, a 10% probability in mCRC and 15% probability in other indications. Our approach assumes a partnering deal or out-licensing of Cavatak in FY19, with costs of subsequent clinical development borne by the partner/licensee.

Viralytics is subject to typical biotech company development risks, including the unpredictable outcome of trials, regulatory decisions, success of competitors, financing and commercial risks. In particular, it has a very high single-product risk, with the entirety of its value residing in Cavatak. The investment case hinges on the outcome of clinical trials and, assuming data are positive, the company’s ability to secure a partnership (or further capital) to advance Cavatak into late-stage trials. Ideally, a partner would have an established oncology franchise with the resources and experience to evaluate Cavatak in multiple cancer indications. The rapidly evolving treatment landscape for melanoma means that the greatest commercial opportunity for Cavatak is likely to be in combination with checkpoint inhibitors or other targeted agents. Initial results from Keynote 200 suggest that combining iv Cavatak with Keytruda leads to high response rates in lung and bladder cancer tumours following iv administration, but only a small number of patients have been evaluated so far. Further results from the Keynote 200 will be important pointers to utility beyond melanoma, although Cavatak can also be administered by ultrasound or CT-guided intralesional injections, as in the upcoming study in mCRC liver metastases.

Viralytics reported cash of A$27.7m at 30 September 2017. The company has the resources to complete ongoing Cavatak combination clinical trials (MITCI, CAPRA and STORM/Keynote 200), plus additional Cavatak combination trials being planned. If a partner is not secured for Cavatak, Viralytics has sufficient resources to initiate a pivotal trial of Cavatak plus Yervoy in melanoma patients who are refractory to ICI therapy, but would need additional funds to complete the trial. Given the positive signs of efficacy to date, prospects for attracting a big pharma partner appear promising.

Viralytics is actively pursuing a clinical trial programme to confirm the potential of Cavatak as an anticancer therapy. The emphasis of the programme is on combination immunotherapy and demonstrating that Cavatak is efficacious when administered intravenously. Exhibit 1 summarises the ongoing and completed Cavatak clinical trials.

Viralytics presented an update at the Society for Immunotherapy of Cancer (SITC) meeting held in Washington DC on 7-11 November 2017 which showed that the MITCI study continues to deliver impressive response rates in patients with PD1-refractory melanoma. MITCI is a Phase Ib trial of intralesional Cavatak in combination with the anti-CTLA-4 ICI Yervoy in patients with advanced melanoma, many of whom had undergone prior ICI immunotherapy.

The preliminary best overall tumour response rate (BORR) in patients who have not undergone prior anti-PD1 therapy (PD1 naïve) was 57% (8/14, see Exhibits 2 and 4), while the response rate in patients who had failed prior single-line anti-PD1 therapy was 29% (2/7, Exhibits 3 and 5).

The preliminary response rates in patients who have failed prior PD1 therapy in the MITCI trial are very encouraging. Exhibit 6 compares response rates in the different patient subgroups in the MITCI trial to response rates reported for comparable populations of melanoma patients treated with Yervoy as a single agent or in combination with other oncolytic virus immunotherapies, with the oral IDO1 inhibitor epacadostat, or with the anti-PD1 ICI dug Opdivo.

The 29% response rate among patients who had failed prior anti-PD1 treatment but had not been previously treated with Yervoy compares favourably to a response rate of only 14% (14/97) when patients who had failed treatment with the PD-1 drug Keytruda were treated with Yervoy on its own, and 11% in Yervoy pivotal studies. Given that anti-PD1 drugs are becoming widely used in first line treatment of advanced melanoma, this class of patients represents a substantial potential market opportunity for Cavatak.

We also note that the 57% response rate in the MITCI trial in PD1 naïve patients is higher than that seen when other drugs were combined with Yervoy in PD1-naïve patients. This includes the 39% (39/98) response rate from a randomised Phase II trial of Yervoy combined with Amgen’s approved oncolytic virotherapy, Imlygic (T-Vec) in which 98% of subjects were PD1 naïve (Chesney et al, ASCO 2017). We note that Amgen is focusing on combining Imlygic with PD1 inhibitors rather than Yervoy in its late-stage development programme.

Although the response rates for Cavatak plus Yervoy are higher than those reported for historical studies, the MITCI data so far includes response data on only 25 patients; the results will need to be further confirmed in the expanded study which will recruit 60 patients in total, with a focus on recruiting additional subjects who have failed prior anti-PD1 therapy.

While bearing in mind that the number of patients who have been assessed to date is small, and that response rates may change as additional patients are assessed, at this early stage it appears that Cavatak is more effective at inducing immune responses in melanoma patients in combination with Yervoy than any other therapy except PD1 inhibitors such as Opdivo, and the Yervoy/Opdivo combination is poorly tolerated by patients.

No Cavatak-related grade 3 or higher adverse events have been reported in the MITCI study, and only four of the 38 patients (11%) who have commenced treatment have experienced Yervoy-related grade 3 or 4 adverse events (six grade 3/4 adverse events in these four patients - fatigue, elevated liver enzymes, pruritus, dehydration, hyperglycaemia).

The 11% treatment-related adverse event (TRAE) rate in the MITCI study is much lower than the 20-27% grade 3 TRAE rate reported for Yervoy on its own in advanced melanoma, and the 55% TRAE rate when Yervoy was combined with Opdivo. The results reported to date suggest that a Cavatak/Yervoy combination is much better tolerated than Opdivo/Yervoy, with similar efficacy. In this regard it is notable that there have been no cases of severe diarrhoea, colitis or pneumonitis, which are a common occurrence with other Yervoy treatment regimens. If the low adverse event rate for the Cavatak/Yervoy combo is confirmed in larger studies, it could give the combo a competitive advantage (if approved).

Viralytics has commenced planning for a pivotal study of intralesional Cavatak plus Yervoy in melanoma patients who have failed treatment with a single PD1/L1 therapy. The trial could commence in Q418 if the combination therapy continues to display durable anti-tumour activity in the MITCI expansion cohort. Based on advice to date the company believes that a study in 200-220 patients randomised to Cavatak plus Yervoy vs Yervoy monotherapy would be appropriately sized. We estimate that cost of such a study would be in the order of US$20m, so it would be within the capacity of Viralytics to conduct this study on its own if it raised additional funding.

However, we believe that under certain circumstances there is the potential that the MITCI data could be used to support an application for accelerated approval in the US. This would depend on continued high response rates in the additional 44 patients (preferably together with continued low adverse event rates) as well as there continuing to be a significant unmet need for an effective treatment for this patient group (in this regard Amgen’s Phase III Imlygic/Yervoy combo trial (NCT01740297) will be an important one to watch).

We note that the FDA approved Bavencio (avelumab) for metastatic Merkel cell carcinoma in March 2017 based on a 33% response rate in an 88-patient, single-arm trial. This data set is not much larger than the one Viralytics will have at the completion of the enlarged MITCI study.

At this stage, it is difficult to tell how realistic a prospect an application for accelerated approval would be; in our forecasts we continue to assume that a Phase III study will be required before Cavatak receives market approval.

Viralytics is testing the combination of intralesional Cavatak with Keytruda (pembrolizumab) in advanced melanoma (Stage IIIB/C and IV) in the Phase Ib CAPRA study. Phase Ib data from the first 23 assessable patients which was presented at SITC in November showed an objective response rate of 61% (4/23), including responses in seven of 11 patients with the most advanced Stage IV M1c disease (Exhibit 9). Importantly, only three of 26 (12%) enrolled patients experienced grade 3/4 TRAEs were observed, all of which were Keytruda related. The trial had enrolled 26 of the target of 50 patients as of early November.

The 61% preliminary response rate for Cavatak plus Keytruda in the CAPRA study is higher than the published rates for either agent used alone in late-stage melanoma: Cavatak 28% and Keytruda circa 33%. Exhibit 10 also shows that the CAPRA response rate is comparable to the 58% response rate in PD1-naïve patients for the approved Opdivo/Yervoy combination, and 57% and 58% for Keytruda combined with Imlygic and epacadostat, respectively, in preliminary studies. Exhibit 10 highlights that the tolerability of Cavatak plus Keytruda was much better than for the other combinations. In particular, Grade 3/4 TRAEs were reported for 55% of patients in the Phase III trial of the approved Opdivo/Yervoy combination, whereas they have only been reported for 12% of patients so far in the CAPRA study, none of which were Cavatak-related.

Epacadostat and Imlygic are both in Phase III studies in combination with Keytruda, with data expected in 2018.

Viralytics also presented initial data at SITC from the expansion cohorts of the Keynote 200 trial, which is being conducted in collaboration with Merck. The expansion cohorts of Keynote 200 are exploring whether viral replication in tumours following intravenous (iv) Cavatak administration can boost the efficacy of Keytruda in lung and bladder cancer patients.

Data was presented on the preliminary assessment of 28 checkpoint naïve patients. Nine of the 28 patients were not evaluable for target lesion response assessment due to early disease progression or study discontinuation. The early disease progression in part reflects the fact that this “all-comers” study is being conducted in a heavily pre-treated patient population, with 28% and 58% of the advanced bladder cancer and NSCLC patients, respectively, having received two or more prior therapies.

Exhibit 11 show that among the 19 evaluable patients responses were observed in three of six (50%) advanced small cell lung cancer (NSCLC) patients and five of 13 (38%) bladder cancer patients. On an intention to treat (ITT) basis, which includes the nine early withdrawals, the response rates were 3/10 (30%) and 5/18 (28%) for lung and bladder cancer, respectively.

While bearing in mind the limitations of cross-trial comparisons and the fact that only a small number of patients have been evaluated to date, the preliminary ITT response rates for iv Cavatak plus Keytruda compare favourably to response rates to Keytruda monotherapy in second line (pre-treated) NSCLC (17-18%) and bladder cancer (21%) in pivotal studies, as shown in Exhibit 13.

Exhibit 12 above shows that 15 days after the start of Cavatak therapy, there was a significant increase in the PD-L1 expression in tumours that were negative or only weakly positive for PD-L1 at the start of the study. The increased PD-L1 expression is a positive outcome because patients with high PD-L1 expression typically respond better to anti-PD1/L1 therapy. For example, in the Keynote-010 study of Keytruda in pre-treated NSLC patients, the response rate in patients with high PD-L1 tumour expression was ~30% vs ~10% for patients with low PD-L1 tumour expression.

So far 64 out of the target of 90 patients have been enrolled in Keynote 200 part B. A further clinical update on the study is expected in Q218.

Viralytics has encountered strong interest in assessing Cavatak in combination with ICI drugs in additional settings and has announced plans to initiate 4 new Phase I trials by June 2018, as shown in Exhibit 14. We expect these studies to generate additional information about the potential breadth of applicability of Cavatak at a relatively low cost – likely around A$1-2m per study. We expect that this information to be of particular interest to potential partners who may be assessing how Cavatak would fit into their own product development pipeline.

Our comments on the rationale for each of the studies include:

Liver metastases in uveal melanoma represent a serious unmet medical need for which there are no effective drug treatments. Uveal melanoma is rare form of melanoma occurring in the eye, representing ~5% of melanoma cases 1. Following local resection of the primary tumour of the eye about half of people will suffer a recurrence, with 90% having liver metastases. Uveal melanomas behave differently to melanomas arising in the skin, and no prospective randomised trial has ever demonstrated an improvement in overall survival for metastatic uveal menanoma1. Response rates to single agent anti-PD1 therapy are only 3-4% in uveal melanoma2. If Cavatak plus the combo therapy can demonstrate efficacy in metastatic uveal melanoma it will be the only treatment to have done so.

This study will investigate whether intralesional Cavatak can increase the response rate to Keytruda, in a similar fashion to the improvements see in the CAPRA study in melanoma. Viralytics has previously treated four HNSCC patients with Cavatak in a Phase I study that ended in 2012. Keytruda is approved as a second line treatment for squamous cell carcinoma of the head and neck (HNSCC), based on a 16% ORR in a study of 174 patients with previously-treated disease.

We note that Amgen’s Imlygic plus Keytruda combo is currently being investigated in the 40-patient Phase Ib component of a Phase Ib/III study in HNSCC3.

The majority of metastatic melanoma patients do not a have a readily accessible lesion for injection because the primary tumour in the skin has previously been excised. iv Cavatak could be a convenient way of treating these patients, and could potentially ensure that Cavatak is accessible for all metastatic melanoma patients.

Viralytics plans to test intralesional Cavatak plus an un-named checkpoint inhibit in patients with microsatellite-stable colorectal cancer (CRC) who have developed liver metastases.

Keytruda and Yervoy are approved for treating a small subset of metastatic colorectal cancer (mCRC) patients, with treatment limited to the 4-5% of mCRC4 that have tumours with a high mutation burden identified as microsatellite instability high (MSI-H) and mismatch repair deficient (dMMR). PD1 ICI have not been shown to be effective in the other 95% of mCRC patients who have microsatellite-stable disease. In studies of PD-1 checkpoint inhibitors in mCRC patients there was only one responder among 33 patients, and that patient was MSI-H5.

CRC is the third most common cancer in the US with 135,000 new cases and 50,260 deaths expected in the US in 2017; globally there were estimated to be 1.4m new cases and 700k deaths in 2012. About 20% of CRC patients eventually develop liver metastases6, for which there are few effective treatments other than surgery. While this appears to be a higher-risk study, we expect that even a single responder would attract considerable commercial interest.

Separately, we note that there is continued interest by big pharma in investigating checkpoint inhibitor combinations in microsatellite stable mCRC, with Roche trialling Tecentriq in combination with its MEK inhibitor Cotellic in this patient group (clinicaltrials.gov NCT02788279).

Viralytics is also undertaking preclinical studies to assess Cavatak in combination with other immunotherapy agents such as LAG-3, TIM-3 and IDO. It has already shown that the triple combination of Cavatak plus an anti-PD1 antibody and an IDO inhibitor significantly reduced overall mouse tumour burden compared to anti-PD1/IDO inhibitor combination.

These studies could add value as Cavatak advances towards a potential licencing, partnering or sale transaction, by expanding its commercial opportunity across a range of disease indications and drug combinations.

We lift our valuation of Viralytics to A$469m or A$1.95/share (undiluted) from A$408m or A$1.70/share. This reflects the following changes:

Our valuation uses a risk-adjusted net present value (rNPV) method to discount future cash flows for the cancer indications shown in Exhibit 12 through to 2044, using a 12.5% discount rate. It assumes a partnering deal or out-licensing Cavatak in 2019 (previously 2018), with the costs of subsequent clinical development borne by the partner/licensee.

Our model includes risk-adjusted upfront payments and clinical, regulatory and sales milestones from a potential licensing deal, based on average Phase II deal metrics from BioCentury (US$25m upfront payment, US$240m total milestones) and our own assessment of the development stage of Cavatak. There is a broad range of value for deals in the oncolytic virus field, from the US$236m Boehringer Ingelheim/Vira Therapeutics deal for a drug that is still in preclinical development and the December 2016 licence deal between Bristol-Myers Squibb and the unlisted British biotech PsiOxus for its preclinical armed oncolytic virus NG-348 (US$50m upfront, and up to $886m in development, regulatory and sales-based milestones), to $1bn ($425m cash upfront and $575m earnout) of the Amgen/BioVex deal for Phase III asset, T-Vec. We maintain our previous assumption that milestone payments for a Cavatak licence deal will total US$355m as the product is generating favourable clinical data, and advancing towards pivotal development.

Viralytics is subject to typical biotech company development risks, including the unpredictable outcome of trials, regulatory decisions, success of competitors, financing and commercial risks. In particular, it has a very high single-product risk, with its entire value residing in Cavatak. The investment case hinges on the outcome of clinical trials and the company’s ability to secure a partnership (or further capital) to advance Cavatak into late-stage trials. Ideally, a partner would have the resources to evaluate Cavatak in multiple cancer indications. The greatest commercial opportunity for Cavatak is likely to be in combination with checkpoint inhibitors or other targeted agents – outcomes of ongoing Phase Ib combination trials could be critical to future clinical and commercial success.

Viralytics reported a loss of A$12.3m in FY16 (A$11.6m when foreign currency translation loss is excluded). R&D expenses for FY17 totalled A$13.5m vs A$8.6m in FY16, reflecting increased clinical development activities. A rebate of A$4.3m was received in Q317 under the Australian government’s R&D incentive scheme, and A$6.5m is expected in FY18. We lift forecast R&D expenditure to A$13.5m in FY18, to match the increased expenditure in FY17. We do not include the foreign exchange translation gain or loss in our financial summary (Exhibit 13). Cash at 30 September of A$27.7m is sufficient to fund operations beyond the end of FY18 in our forecasts. We model a partner funding a significant proportion of R&D costs in FY19, but do not included any upfront or milestone payments from a potential licensing deal in FY19 in our financial forecasts.