Onxeo — Green light for Ph Ib/II after DRIIV-1 interim data

The open-label, dose escalation Phase I DNA Repair Inhibitor administered IntraVenously (DRIIV) recruited patients with various advanced solid tumours, with the first patient treated in April 2018. This is the most advanced study where patients receive AsiDNA systemically (positive findings from the Phase I trial with topical administration of AsiDNA have already been published). The study aims to assess the safety/tolerability profile and recommended dose for subsequent efficacy trials. Onxeo also explored various biomarkers that might help to gauge the activity of AsiDNA and stratify patients in later trials.

As planned, Onxeo released interim results from the first part of the trial on 5 November 2018, which included results from the first three dose groups. The findings from a total of 10 patients who received 112 infusions of AsiDNA ranging from 200mg to 600mg include:

The rationale for this study was built on previous findings from the Phase I DRIIM, where AsiDNA has been administered intratumourally in melanoma patients, as well as extensive preclinical work to determine activity biomarkers. While the data released from the DRIIV-1 trial are still very early and no conclusions about efficacy can be made yet, we find it reassuring that no serious drug-related side effects emerged with intravenous administration. AsiDNA is a unique drug (DNA damage repair inhibitor) with no close comparators with respect to mode of action and therefore there was no visibility on the safety/tolerability profile via a systemic administration before DRIIV-1 results (safety data via local injection in the DRIIM trial was good). The consistent pattern of increase in activity biomarkers also demonstrated that the drug reaches its target after the intravenous administration.

With the results announcement, Onxeo also provided an update on its AsiDNA development strategy. The company aims to initiate Phase Ib/II studies in H119 with AsiDNA in combination with other established treatment options, primarily PARP inhibitors and standard-of-care (SoC) chemotherapy regimens. Specific indications have not been announced yet, but since the acquisition of the asset in March 2016, Onxeo has conducted various preclinical studies to understand how to better position AsiDNA in the clinic.

Single strand DNA breaks are repaired by base excision repair (BER) pathway. Among other enzymes involved in BER are poly(ADP ribose) polymerase 1 and 2 (PARP1 and PARP2), which act as sensors and signal transducers. PARP inhibition therefore affects this pathway specifically. PARP inhibitors have shown promising efficacy and safety in clinical trials with BRCA mutated tumours, but the main drawback was the necessity of a dysfunctional homologous recombination (HR) pathway (cells can compensate DNA repair via this pathway). This underlies the synthetic lethality mechanism of action exhibited by PARP inhibitors, since those cancers with BRCA mutation (ovarian, breast) are not able to repair DNA damage by HR pathway. However, because of the dependence on dysfunctional HR pathway, the studies have shown that even those tumours that were initially responsive to PARP inhibitions, finally relapsed through compensatory mutations restoring the HR activity.

First-in-class AsiDNA is based on signal-interfering DNA technology, which if introduced into a cell acts as a signal mimicking the damage of the cell’s own DNA. AsiDNA molecules are short double-strand DNA that mimic double-strand breaks and are recognised as “damaged DNA” by repair and signalling proteins. Namely, AsiDNA hyper-activates PARP1 and the DNA-dependent protein kinase (DNA-PK) among others leading to a cascade of repair proteins being recruited to “repair the damage”, as a result of which the actual damage of a cell’s DNA remains unrepaired. This action renders the HR and NHEJ pathways dysfunctional. Due to its independent mechanism of action (abrogation of the HR pathway) there is strong rationale to use AsiDNA in combination with PARP inhibitors to potentiate their effect. Inhibiting the BER pathway through PARP inhibition, in addition to HR/NHEJ pathway inhibition with AsiDNA, will cause accumulation of double-strand DNA damage in the tumour cells and tumour cell death. In addition, AsiDNA could potentially be used to sensitise BRCA non-mutated tumours to PARP inhibitors, which in turn would expand their use substantially.

In our update note, we reviewed the preclinical in vitro data Onxeo presented at the AACR annual meeting in April 2018, which showed the synergistic effect of AsiDNA in combination with the company’s HDAC inhibitor, belinostat. Onxeo also obtained evidence that AsiDNA may induce ‘autosensitisation’, ie the tumour cell’s sensitivity to AsiDNA could actually increase as the treatment progresses, unlike the majority of drugs in oncology that develop resistance. Previously, Onxeo also showed that low doses of AsiDNA in combination with classical chemotherapy agent carboplatin outperformed carboplatin alone in a triple negative breast cancer (TNBC) model.

On 12 July 2018, Onxeo announced another set of positive preclinical data with AsiDNA in combination with various PARP inhibitors. AsiDNA was tested with olaparib (Lynparza, AstraZeneca, consensus estimate of $621m in 2018 sales) and talazoparib in in vivo and in vitro models. The findings suggest the following:

Ovarian cancer and breast cancer are the only approved indications for PARP inhibitors so far. These cancers were seen as the low-hanging fruit for PARP inhibitor development, due to the link between BRCA mutation and these tumours (9% of breast cancer, 11-15% ovarian cancer is caused by BRCA1/2 mutations). The market for PARP inhibitors is currently around $1bn worldwide (source: EvaluatePharma), with ovarian cancer (Lynparza, Zejula and Rubraca) and breast cancer (Lynparza, talazoparib) driving near-term sales (Exhibits 4 and 5). According to consensus, the market for PARP inhibitors could reach $6.9bn in 2024. This will be driven by the introduction of new PARP inhibitors, possible transitioning of 2L/3L products into 1L, new indications (where prostate is expected to perform best) and combinations with other drugs. Although further indications are being explored, ovarian cancer and breast cancer are expected to remain the largest indications for PARP inhibitors into 2024. Onxeo is hoping to move into this growing market by developing AsiDNA + PARP inhibitor combinations(s) in solid tumour indications, which makes sense due to the complementary mechanism of action, but also due to the large market opportunity. In addition, AsiDNA + PARP inhibitor combination could be used regardless of the mutation status, which would have the potential to significantly expand indications for PARP inhibitors themselves.

PARP inhibitors are the only marketed class of DNA repair inhibitor, but drugs targeting other DNA repair pathways and targets are also in clinical development. Large pharma are very active in the space, especially AstraZeneca and Merck KGaA, where AstraZeneca is developing its Lynparza franchise against PARP (Exhibit 6), but also investigating ATM (HR pathway), DNA-PK (NHEJ pathway) and WEE1. Merck is also very active in the space, developing drugs against PARP, but also ATR (HR pathway), ATM (HR pathway), DNA-PK (NHEJ pathway) and WEE1. As far as we are aware, AsiDNA is differentiated and has a unique mechanism of action as it is an agonist that acts as a decoy and affects multiple proteins in the DDR cascade, while other drug candidates in the area act as inhibitors of specific single targets.

On 7 June 2018, Onxeo announced that it had effectively sold the Beleodaq royalties from its partnership with Spectrum Pharmaceuticals to SWK Holdings Corporation in exchange for an immediate payment of $7.5m. SWK Holdings Corporation is entitled to receive $13.5m in future royalties from Spectrum. Our last published rNVP of Beleodaq (marketed) was €34.5m, which was based on a bottom-up model. Onxeo reported €1.0m in licensing income in H118 alone. We believe its future commercial potential of Beleodaq is higher than $13.5m. In such a case Onxeo would retain the residual royalties or those could be sold again. Therefore, to reflect both the received payment and the remaining value of the asset, we subtract $7.5m (€6.6m, the price paid by SWK for Beleodaq royalties) from Beleodaq’s rNPV in our model and add to the cash balance. We see the deal as neutral to the company’s value, however, it provided a decent increase in cash to help fund Onxeo’s ongoing development programmes and signals the prioritisation of the AsiDNA programme.

As a next step, we reviewed the potential of the products since Spectrum Pharmaceuticals reported several quarterly sales data points after our last revision. Beleodaq sales in the US in Q318 were $3.2m vs $3.4m in Q317. 9M18 sales were $8.6m vs $9.7m in 9M17. This was likely due to istodax (Romidepsin, Celgene, sales of $76m in 2017), which is also indicated for peripheral T-cell lymphoma (PTCL), going off patent and the first generic forms appearing at the beginning of this year. Consensus expects that Romidepsin will still bring in sales of $51m by 2024 (source: EvaluatePharma), although it is indicated for PTCL and cutaneous T-cell lymphoma. Our previous estimate for Beleodaq FY18 US sales was $20.5m, which we now revise to $11.5m. We have also revised our peak penetration rate from 15% to 7.5%, which translates into lower peak sales expectations of $40m for Beleodaq (US, PTCL second line) and $30m for Beleodaq (EU), and lower respective NPVs. From a strategic perspective, variations in end-user sales are of less importance for Onxeo now, as the royalty sale deal ensured cash up front. The main value driver in our model is AsiDNA (Exhibit 8).

We have also decided to remove Livatag from our valuation. The results released a year ago (September 2017) showed that the primary endpoint of the Phase III trial with Livatag was not reached. We reduced the success probability to 5% as Onxeo was still considering all options for the asset but, given there were no recent updates to this end, we conservatively remove it from our valuation.

On 15 June 2018, Onxeo announced that it had obtained an equity financing line from Nice & Green by issuing new shares over a 10-month period for a maximum amount of €5.4m. The shares will be issued based on the average of the volume weighted average share price of the three trading days preceding each issue, minus a maximum discount of 5.0%. The reported cash position at end-Q318 was €13m, which includes proceeds of €2.4m from drawdown of the equity line. According to the company, this extended the cash reach into 2020, which is in line with our model. Other operating estimates were only fine-tuned.

Our valuation is €172m or €3.3/share, lower than our previous valuation of €221m or €4.4/share, mainly due to the R&D revisions as detailed above, which were offset by a higher cash position and rolling our model forward. Our assumptions for AsiDNA are unchanged although, as Onxeo progresses its preclinical development into further trials, there is potential to add more indications for AsiDNA as monotherapy or in combination with other agents such as PARPi or chemotherapy drugs.