Targovax — ONCOS-102 set for late-stage development

On 22 January 2020, Targovax reported randomised data from the Phase I/II study (n=31) in unresectable malignant pleural mesothelioma. The randomised, open-label trial compared ONCOS-102 plus standard of care (pemetrexed/cisplatin) versus standard of care treatment in first- and second-line settings. The primary goal of the study was to evaluate the safety and tolerability of ONCOS-102, which is typical for this stage. Key secondary endpoints will evaluate:

In addition to the overall results, first-line and second-line (or later) patient subpopulations were also evaluated.

The results confirmed a good ONCOS-102 safety profile, ie the combination with pemetrexed/cisplatin did not show additional toxicity beyond that expected with the standard pemetrexed/cisplatin chemotherapy. The clean safety profile will be especially beneficial in subsequent studies, where Targovax expects to test ONCOS-102 in triple combination in mesothelioma (anti-PD1/L1 plus chemotherapy plus ONCOS-102). Both standard-of-care chemotherapy in mesothelioma and CPIs have well documented toxicity profiles. Given the advent of CPIs, the standard of care is likely to change in mesothelioma treatment to include CPIs. In such settings, ONCOS-102’s benign safety profile could prove invaluable.

We note that descriptive patient data (Exhibit 2) showed that the experimental arm had generally more advanced patients, and there was therefore a relatively higher hurdle rate for ONCOS-102 to demonstrate efficacy against the control.

The key clinical response endpoint of mPFS was 8.4 months (active arm) vs 6.8 months (control) and above the historical control of 5.7–7.3 months. The mPFS results are still early, with many patients not included in the analysis (additional data will serve as a potential catalyst later in 2020). In first-line patients, mPFS was 8.9 months vs 6.8 months (Exhibit 1). Data for the second-line group were still not mature enough for analysis.

ORRs in first-line patients were 30% in the experimental arm vs 33% in the control arm (the experimental arm had more advanced patients). ORRs in second-line patients were 11% in the experimental arm (n=9) vs 60% (n=5). The response in the control arm is clearly unusually high and likely the result of a small sample. This negatively skewed analysis of the ORRs in the overall population. Still, these ORR data compare well to historical controls of c 20% (Exhibit 3).

One trial (Vogelzang et al, 2003) reported an ORR of 41%, but the FDA later queried this. This large Phase III study (n=456) was the basis for approval of pemetrexed’s addition to cisplatin, which became standard of care in early 2000s. The investigators concluded that in chemotherapy-naive patients who were not eligible for curative surgery, pemetrexed plus cisplatin demonstrated mPFS of 5.7 months versus 3.9 months in the cisplatin alone arm. Reported response rates were 41% and 17% in the respective arms. A review conducted by the FDA later concluded that there was considerable discrepancy in tumour response evaluation among the investigators and the FDA reviewers (Hazarika et al, 2005). This was attributed to the fact that because of the nature of the tumour, which often grows in diffuse sheets rather than spherical configurations in pleural cavity, radiographical evaluation can be complicated. The FDA could agree to only 47 out of 94 patients in the pemetrexed plus cisplatin arm, who were reported by the investigators as showing a response to treatment. However, the FDA concluded that tumour responses were more frequent in the pemetrexed plus cisplatin arm, therefore the drug was approved, but without stating specific tumour response or time to tumour progression numbers. The response rate comes down to 21% if calculating the response rate using the number of patients the FDA agreed on (Exhibit 3).

Tumour biopsy results showed that 10 out of 15 evaluable patients in the experimental group had increased tumour-infiltrating CD8+ T-cells. 9 of these 15 patients had increased PD-L1 expression in the tumour. 7 of these 9 were still progression free at the time of the analysis. While the control group was small, infiltration in the active arm was very pronounced in some cases, indicating ONCOS-102 activity, in our view. These immunological data in combination with the clinical response show that the mechanistic effect of ONCOS-102 is potentially associated with a clinical improvement. We also note that in this particular trial the patients received chemotherapy and immunotherapy (ONCOS-102). Immunosuppression (or myelosuppression) is a known side effect of chemotherapy, especially in cisplatin’s case. So, the fact that ONCOS-102 generated strong immune response in such a setting is encouraging. Targovax should present more biomarker data in due course.

While the size of the study limited the analysis for statistical significance (the trial was not designed to check for a statistically significant clinical effect), the encouraging mPFS result and the fact that clinical outcomes seem to correlate with a cancer-specific immune response are clearly positive signals. We believe the data support the increased probability of approval. In our model, we have increased the success probability from 10% to 20% (standard industry data, Wong et al, 2019; Biotechnology Innovation Organization, 2016).

Targovax appears to have a clear plan to advance ONCOS-102 testing in mesothelioma to late-stage development. While, understandably, the concrete trial design and timelines are not yet available, the likely positioning of ONCOS-102 and the preliminary study plan are clear (Exhibit 5). The next trial will assess ONCOS-102 in a triple combination including standard chemotherapy and CPIs. In addition, Targovax indicated that it is already in discussions with a prospective pharma partner for a study collaboration, although the extent of the possible partnership is not clear yet.

While the current standard of care is still chemotherapy, CPIs are increasingly being used as a salvage therapy. Based on our review of literature (eg Nowak et all, 2019; Forde et all, 2019; Bibby and Maskell, 2019) and discussion with Targovax’s management, we believe that CPIs plus chemotherapy will become a new standard of care in the near future. Therefore, the intention to focus on triple combination with ONCOS-102 makes sense from a strategic point of view (there were many cases in the past where a biotech company using contemporary standard of care conducted a successful trial only to find that the standard of care had changed, rendering all data virtually useless). Existing data also support the rationale to combine CPIs with ONCOS-102 (specific immune system activation, increased PD-L1 expression). As discussed above, we believe the strong safety profile of ONCOS-102 will be key in triple combination trials, as both chemotherapy and immunotherapy with CPIs have distinctive toxicities.

Overall, as we described in our last note, mesothelioma is one of the most difficult cancers to treat, with classic chemotherapy still being the standard of care. There is a clear unmet need in this indication, given the aggressive nature of the cancer and a lack of innovative treatment options. Although plans for the next stage are still preliminary, we believe an accelerated approval is possible in this indication. Pemetrexed was approved based on a single trial (Vogelzang et al), as the FDA acknowledged that it was a large trial in a challenging indication, which is unlikely to be replicated (Hazarika et al). Targovax has also indicated that the next trial could become a pivotal programme.

Our valuation has increased to NOK1.5bn or NOK19.9/share from NOK1.2bn or NOK18.7/share after increasing the success probability in our mesothelioma projects to 20% from 10%. We also included the private placement of NOK101m gross, which had a dilutive effect and partially offset the increased valuation per share. Our other assumptions (including R&D costs) remain unchanged for now, however, we will revise our mesothelioma project again once Targovax presents more details. Upcoming Phase I melanoma trial results and more data from the Phase I/II mesothelioma trial are potential catalysts in the near term.