Investment case

Redx Pharma is a clinical-stage biopharma focused on oncology and fibrosis that aims to identity and develop “best-in-class” or “first-in-class” small molecules with the emphasis on scientifically validated targets. Lead candidates will be progressed to Phase II proof-of-concept before strategic options are assessed; however, other assets can be, and have been, out-licensed earlier if proposed returns are sufficiently attractive. Redx listed on AIM in 2015, raising £15m; with a further £10m in 2016, £12m in 2017, and £25.7m in December 2020. Other financing events include: MGL capitalising its £2.5m 2019 loan in 2020; share subscriptions by Redmile (£1.3m in March 2020) and Sofinnova (£0.8m, July 2020); and $29m in convertible loan notes issued to the latter two investors (July 2020). In December 2020, c20% of the CLNs were converted, with £17.1m now outstanding. Polar Capital also joined the share register in December 2020. To date partnered assets have contributed c £35m in milestones and fees. Redx is headquartered near Manchester (in Alderley Edge, Cheshire), and has c 70 full-time employees.

Valuation

We value Redx Pharma using an rNPV of the known development programmes together with an estimate of the inherent worth of the now well-proven drug discovery platform. These are netted out against operational costs and net cash. Success probabilities are based on standard industry criteria for each stage of development but flexed to reflect the characteristics of the differing indications. We employ conservative assumptions throughout, erring on the side of caution with the timing of clinical trials, market launches, adoption curves, and patient penetration. The next 12-18 months should bring material news flow and opportunities for value inflection. Our valuation is £434m, or 157.5p/share.

Financials

The cash position at end-FY21 (30 September 2021) was £29.6m (end-FY20: £27.5m) having been boosted by £25.7m (gross) proceeds from the Placing and Open Offer in December 2020, plus milestone receipts of £2.8m ($4m) and £2.2m ($3m) received during the period from AstraZeneca and Jazz Pharmaceuticals. Post-period end, a further £7.3m ($10m) was received from Jazz Pharmaceuticals and £6.5m ($9m) from AstraZeneca as key milestones were achieved. The level of investment in RXC004 and RXC007, as they continue into and through Phase II, means current cash provides a runway through Q4 calendar year 2022 (Q1 FY23).

Sensitivities

Redx Pharma seeks to either address validated molecular targets and create a “best-in-class” or, if a target is particularly attractive, to create a “first in class” molecule. Arguably, developing small molecules is less risky than an equivalent biological compound. Irrespective of the merits of this point, the typical industry risks associated with clinical trial results, navigating regulatory hurdles, ensuring sufficient financing is in place, partnering discussions and, eventually, pricing and commercialisation still apply.

Redx Pharma: Ambitious and focussed

Redx Pharma has a proven discovery platform, underpinned by the strength of its medicinal chemistry expertise. An ambitious and capable management is executing a clear and focussed strategy that aims to create either best-in-class or first-in-class small molecules addressing clear needs in oncology and fibrotic diseases. The pipeline is well-balanced; two in-house programmes (RXC004 and RXC007) are progressing well, two other candidates have been successfully partnered, and earlier stage assets are showing promise. RXC004, a Porcupine inhibitor, is in Phase II trials in Wnt ligand driven tumours and RXC007, a ROCK2 selective inhibitor, should start Phase II studies in 2022. Although continued progress of RXC004 and RXC007 will inevitably attract investor attention, we believe the whole business (scientific, operational, and personnel aspects) is now set to generate material value enhancement. Our rNPV based valuation of Redx, with conservative assumptions, is £434m or $564.7m.

Redx is increasingly recognised as a creator of innovative and highly differentiated small molecule drug candidates. It has established an industry-wide reputation for its medicinal chemistry expertise, which underpins its discovery activities. Given its size, research productivity has been impressive, both in quality and quantity: four Redx-discovered candidates have entered the clinic, including most recently AstraZeneca’s AZD5055 (formerly RXC006). Redx is focused on oncology and fibrosis, where there are multiple large and attractive indications that are poorly served with existing therapies. The aim is to address validated receptors as well as novel targets, generating “best-in-class” or “first-in-class” programmes that are progressed quickly to key evaluation points, with rapid “go/no go” assessments.

The quality of the science is complemented by an impressive management team, which brings big pharma professionalism and expertise. Swift decision making and development speed is tempered with active risk minimisation. Out licensing of RXC006 in 2020 is a pertinent example; this commercially attractive deal brings in funds and reduces the weight of the Porcupine inhibitor class as a development risk. The Redx pipeline is well-balanced, with two clinical in-house programmes and two late preclinical/early clinical assets out-licenced (to AstraZeneca and Jazz Pharmaceuticals). In-house discovery effort is focussed on highly selective small molecules, directed at known and scientifically validated pathways, for the treatment of genetically defined tumours and poorly treated fibrotic diseases.

Management has delivered material progress across its in-house and partnered programmes. RXC004, an innovative Porcupine inhibitor for genetically selected oncology indications, is in Phase II trials, with monotherapy arms underway and a checkpoint inhibitor combination arm starting shortly. These studies will evaluate efficacy and safety in Wnt ligand driven tumours, notably selected microsatellite stable (MSS) metastatic colorectal cancer (mCRC), selected pancreatic cancer, and unselected biliary cancer. RXC007, a ROCK2 selective inhibitor for various fibrosis indications, is set to initiate a Phase II trial for idiopathic pulmonary fibrosis (IPF) during 2022. If successful, a 12-month Phase IIb trial will evaluate RXC007 plus standard of care (SoC) with forced vital capacity (FVC) lung function as a primary endpoint. Other fibrotic indications could also be explored. The next 12-18 months should generate sizeable news flow that will, if positive, shift investor perceptions. We currently value Redx at £434m ($564.7m) or 157.5p/share.

Differentiated by medicinal chemistry expertise

Redx Pharma’s proven medicinal chemistry capabilities underpin our investment case. The current pipeline has a clear focus on highly selective small molecules for the treatment of genetically defined tumours and poorly treated fibrotic diseases. This proprietary development portfolio (Exhibit 1) is well-balanced, including two clinical stage in-house programmes (a Porcupine and a ROCK2 selective inhibitor), two earlier stage assets that are partnered with AstraZeneca (also Porcupine) and sold to Jazz Pharmaceuticals (pan-RAF), and several research-stage programmes with Redx targeting three further wholly-owned INDs by 2025. The pipeline is sourced from the in-house discovery platform which aims to create both first-in-class molecules that address novel and biologically relevant targets and best-in-class drugs that are directed at already proven and characterised targets.

Such productivity can rightly be viewed as confirmation of Redx’s prowess in the discovery stages, with the quality of the partnering deals showing the commercial relevance of these programmes. Yet, somewhat ironically, the first tangible validation of the approach was demonstrated by a $40m (£30.2m) deal struck with Loxo Oncology in 2017 for a Redx-created Bruton’s tyrosine kinase (BTK) inhibitor, that was ready to enter clinical development. This became LOXO-305 (now pirtobrutinib) and, interestingly, when Eli Lilly acquired Loxo in 2019 for $8bn, LOXO-305 was cited as one of the key desired programmes. Pirtobrutinib is currently in multiple Phase III trials for various haematological cancers. A rolling NDA submission with the FDA seeking accelerated approval in the lead indication, mantle cell lymphoma, is underway; an approval decision is expected early 2023.

Exhibit 1: Redx Pharma pipeline

Source: Redx Pharma   Note: DDR = discoidin domain receptor; GI = gastrointestinal; IPF = idiopathic pulmonary fibrosis; MAPK = mitogen-activated protein kinase; MSS mCRC = microsatellite stable metastatic colorectal cancer; RAF = rapidly accelerated fibrosarcoma; ROCK = Rho associated protein kinase

Redx’s investment case centres on three value pillars: lead assets RXC004 and RXC007, and the discovery engine and expertise embedded within it. Both RXC004 and RXC007 have already delivered compelling albeit early data, with further milestones expected in the near-to-midterm. While the discovery engine is the source of current pipeline assets and existing partnerships, it is now starting to become increasingly visible with disclosures about likely future outputs. Two programmes have been highlighted that target potentially high value and high unmet need disease areas. There is also the promise of updates on the partnered programmes, although these are largely outside of Redx’s influence. While COVID-19 restrictions remain a sensitivity with respect to timings (impacts on patient recruitment into clinical trials is a known industry-wide consequence), we anticipate Redx will make significant strategic progress. Exhibit 2 illustrates expected news flow and catalysts over calendar 2022.

Exhibit 2: Multiple potential milestones in 2022

Source: Redx Pharma

Redx’s clinical stage pipeline is expected to advance and expand over the coming years. We note that ahead of this the company has further strengthened its executive team, with the notable addition of Dr Jane Robertson as Chief Medical Officer, and made several changes to the board with the appointment of Dr Jane Griffiths (formerly Global Head of Actelion and Company Group Chair of Janssen EMEA) as Chair and Dr Rob Scott (formerly CMO of Abbvie and VP Global Development at Amgen) as Non-Executive Director. These hires and appointments bring significant highly relevant development, regulatory, and strategic expertise as Redx builds its clinical portfolio.

RXC004: in Phase II proof-of-concept trials

RXC004 is a Porcupine inhibitor under evaluation as both monotherapy and in combination with checkpoint inhibitors (CPIs) in various solid genetically selected tumours. Porcupine (Porcn) is a membrane-bound enzyme (MBOAT) that enables a key step required for the secretion, transportation, and activity of Wnt ligands. The Wnt (Wingless type) signalling pathways are increasingly recognised as an attractive, albeit challenging, drug target. Gradients of diverse Wnt proteins regulate fundamental processes such as cell development, renewal, and differentiation, hence their important roles in oncology and fibrosis indications.

Wnt signalling has been known as a key oncogenic pathway in multiple cancers for nearly 40 years, with clinical progress hampered by the sheer complexity of the cascades, their key role in so many biological functions (they are involved in virtually all aspects of embryonic development and in control of homeostasis in adult tissues), and the need to better identify suitable patient and tumour types for treatment.

Wnt pathway has been a key drug target for decades

Wnt is linked to a wide range of conserved biological processes with Wnt signalling activating what can be considered, rather simplistically, two distinct pathways: the canonical (β-catenin) and non-canonical (non-β-catenin) systems. These diverse cascades impinge on multiple developmental decisions through the control of cell-cell communication and play a key role in tissue homeostasis and repair. It is now clear that Wnt ligands play a critical role in balancing cell proliferation, differentiation, and cellular homeostasis and dysregulation is a driver in the development of many solid tumours, notably CRC (colorectal cancer), and, more broadly, in a cancer cell’s immune evasion (Exhibit 3).

Exhibit 3: The Wnt signalling pathway

Source: Redx Pharma

The canonical cascade is a key regulator of normal development and tissue homeostasis, and its disruption underlies the initiation and progression of multiple different tumour types. The β‐catenin‐independent (non-canonical) cascade is similarly complex, with new interactions identified as academic research in the Wnt field gathers pace. The two notable non-canonical pathways can be classified as planar cell polarity (PCP) and Ca²⁺ signalling. Both canonical and non-canonical cascades are also implicated in cancer progression: they are known to impact the tumour micro-environment (TME) to promote tumour cell survival and therapeutic resistance (eg through PI3K-AKT).

Porcupine inhibition decreases Wnt ligand signalling

Porcupine sits at a strategic node, with all 19 secreted Wnt family members dependent upon the same biosynthetic enzymes to supply a single fatty acid adduct (palmitoylation) that is required to enable their transport, secretion, and activity. Following palmitoylation, Wnt ligands are secreted to form various Wnt receptor complexes, with RNF43 (Ring finger protein 43) and ZNRF3 (zinc and ring finger protein 3) being key regulators. The RSPO (R-spondin) family is an inhibitor of RNF43/ZNRF3, so deactivating mutations in RNF43/ZNRF3 or activating RSPO fusions can increase levels of the Wnt receptor complex and lead to increased ligand-dependent signalling. Hence, tumour cells that are dependent on Wnt ligands would be sensitive to Porcupine inhibition.

The broad influence of the Wnt pathways means systemic blockade of Porcupine and, subsequent global elimination of Wnt signalling, is unwelcome. The actual effect of the Porcupine enzyme is complex and dependent on the cell type and the physiological context. For any therapeutic inhibition, a key concern is avoiding undesired, yet still on-target, effects (such as on gastrointestinal homeostasis and bone metabolism). The key is careful target selection and then to establish, and operate within, an appropriate therapeutic window (hence the importance of the RXC004 preliminary Phase I results discussed later).

Targeting appropriate cancer types is important. Aberrant Wnt pathway activation is typically achieved through the upstream loss of function (LoF) mutations in RNF43 or RSPO2/3 fusions; both types of mutation result in the upregulation of ‘frizzled’ Wnt ligand receptors, so promoting Wnt signalling. Encouragingly, Porcupine inhibition exquisitely impacts the growth of tumours that depend on Wnt ligand signalling. Such mutations are found in many solid tumours, especially in difficult to treat colorectal, biliary tract, and pancreatic cancers. In addition, head and neck squamous cell carcinoma (HNSCC), squamous non-small cell lung cancer (NSCLC), and castrate resistant prostate cancer (CRPC) cell lines carrying related mutations appear particularly sensitive to Porcupine inhibition.

As mentioned, there is evidence of a strong correlation between Porcupine inhibition and immune tolerance within the tumour micro-environment. Wnt signalling plays a vital part in immune cell modulation, development, activation, regeneration, and downregulation, with Wnt ligands having an essential, and considerable, role in the regulation of immune cells such as dendritic cells (DCs), natural killer (NK) cells, T cells, macrophages, and B cells. It is now established that heightened intra-tumoural Wnt signalling correlates with tumour immunomodulation and immune suppression, which likely contribute to the decreased efficacy of multiple cancer therapeutics. Hence, interacting with Porcupine pathways may result in both a direct effect on tumour cells and indirectly through facilitating improved immune responses; for instance, preventing emerging resistance to a checkpoint inhibitor (CPI) or enabling existing tolerance to be overcome (turning a “cold” tumour “hot”).

RXC004 Phase I monotherapy module at ESMO 2021

RXC004 is a highly selective and potent Porcupine inhibitor which, in preclinical studies, was shown to have promising direct anti-tumour activity in Wnt ligand driven cancer lines and to enhance the immune response in the tumour microenvironment. Redx is studying RXC004 both as monotherapy and in combination with checkpoint inhibitors in various solid tumours.

The RXC004 Phase I programme consists of three modules in an all-comers population each evaluating a different setting: Module 1 is monotherapy with continuous dosing; Module 2 is in combination with the PD-1 checkpoint inhibitor nivolumab (Opdivo, Bristol Myers Squibb); and Module 3 will explore intermittent dosing schedules. Successful outcomes have already supported the initiation of Phase II monotherapy proof-of-concept studies in patients with Wnt ligand driven tumours. First clinical results from this programme presented at ESMO 2021 confirmed that RXC004 has a useful therapeutic window, is safe and well tolerated at the selected Phase II dose, and is applicable to genetically selected Wnt ligand driven tumours given early indications of activity in such tumours.

Exhibit 4: RXC004 Phase I dose escalation design

Source: Redx Pharma, ESMO (European Society for Medical Oncology) Natalie Cook MD PhD

In this first Phase I clinical trial, Module 1 (dose escalation of RXC004 monotherapy) reported results at ESMO 2021. This study module enrolled 25 patients across five centres in the UK, with 24 patients at five dose levels ranging from 0.5mg to 3.0mg. The first patient was treated with a 10mg dose which was not tolerated. Patients enrolled in the study had unselected advanced solid tumours and had received a median three prior lines of therapy, with no standard treatment options remaining. The primary endpoints were safety and tolerability, alongside pharmacokinetics (PK) and pharmacodynamics (PD), and evaluation of preliminary indications of efficacy.

The key finding is RXC004 is safe and well tolerated at the selected Phase II monotherapy dose. Treatment related adverse events (Exhibit 5) were in line with the expected profile for Porcupine inhibition, and dysgeusia (loss of taste) was the only one that was dose related. Dose limiting toxicities (DLTs) were seen in four patients: unsurprisingly, these were most extensive in the 10mg patient.

Importantly, effects on bone metabolism would be expected at the higher doses but prophylactic administration of denosumab (Prolia/Xgeva, Amgen), a RANK ligand inhibitor that decreases bone breakdown, successfully prevented disruption of metabolism and incidence of spontaneous fractures. Gastric events (eg colitis) would also be expected at higher doses but this was not seen in patients who started on 2mg or lower. It will be interesting to see if colitis occurs when used in combination with a PD-1 inhibitor given that this is an observed side-effect of PD-1 therapy.

Exhibit 5: RXC004 monotherapy Phase I treatment-related adverse events

Source: Redx Pharma, ESMO (European Society for Medical Oncology) Natalie Cook MD PhD  Note:  All treatment related adverse events (any grade) occurring in at least 20% patients

In terms of dose selection, the 2mg RXC004 dose struck the ideal balance, achieving an attractive PK/PD profile, with once-daily dosing maintaining levels above the activity target ranges derived from preclinical studies, with containable adverse effects. This once daily dosing profile appears to differentiate RXC004 from other Porcupine inhibitors which have half-lives supporting either twice daily (Novartis’ WNT974) dosing or dosing every other day (A*STAR’s ETC-159). The collective results from these three Porcupine inhibitors from their clinical trials to date suggest this approach to Wnt ligand inhibition is safe and viable.

Data presented at ESMO also showed early signs of efficacy. We note that the patients enrolled were not selected for Wnt ligand driven tumours; however, this was retrospectively analysed. In total, 18 patients had RECIST evaluable disease, of which seven patients had Wnt ligand dependent tumours, six had Wnt ligand independent tumours, and the remaining five had tumours of unknown status. With the caveats that the study was small (18 patients with RECIST evaluable disease), the patients were not genetically selected, and they were also heavily pre-treated, there were encouraging signs of stable disease.

Five of the seven Wnt ligand dependent patients showed durable disease control (Exhibit 6). In contrast, the eleven patients with unknown/Wnt ligand independent disease showed progressive disease. Median treatment duration for the seven Wnt ligand dependent tumours was 13.1 weeks (range: 8.4 to 25.4 weeks) versus 6.6 weeks (5.4 to 7.3) for the unknown or Wnt ligand independent tumours.

Exhibit 6: Clinical activity by Wnt ligand dependence group

Source: Redx Pharma, ESMO (European Society for Medical Oncology) Natalie Cook MD PhD

Study conclusions are summarised in Exhibit 7. Again, with the standard caveats that this is a small study and comparisons across different studies are fraught with difficulties, the data from both this trial and WNT974’s Phase I (Rodon) study (see later) are consistent with differential activity in Wnt-ligand dependent cancers.

Exhibit 7: Summary of Phase I study of Porcupine inhibitor RXC004

Source: ESMO (European Society for Medical Oncology) Natalie Cook MD PhD

Module 2 of the Phase I study of RXC004 in combination with PD-1 checkpoint inhibitor nivolumab is underway. Its primary aim is to evaluate the safety and tolerability of the combination in patients with unselected advanced malignancies; although, as in the monotherapy Module 1, some efficacy signals may become apparent. So far, a 1mg RXC004 dose + a standard nivolumab dose has been shown to be well tolerated. Dosing of RXC004 1.5mg + nivolumab is ongoing, and results of this will determine whether escalation to RXC004 2mg (monotherapy dose) is merited. A Phase I study of Novartis’ WNT974 in combination with its anti-PD-1, spartalizumab, showed an acceptable safety profile (even though 75% of patients experienced a treatment related adverse event). Interestingly, although the patients were not genetically selected, there were promising signs that “blocking Wnt signalling may enable response to checkpoint inhibition”.

Pivotal Phase II proof-of-concept trials ongoing

The RXC004 Phase II programme similarly explores both monotherapy (patient enrolment started in Q421) and a CPI combination with nivolumab (expected to begin during H122, COVID restrictions permitting). Two multi-arm Phase II studies are planned: PORCUPINE in genetically selected MSS mCRC includes a RXC004 monotherapy arm and a combination arm with anti-PD1 agent nivolumab; and PORCUPINE2, a monotherapy trial with two arms, one in genetically selected metastatic pancreatic cancer and the second in unselected biliary cancer. Exhibit 8 provides an overview of the Phase II programme.

Exhibit 8: Phase II RXC004 programme overview

Source: Redx Pharma

The Phase II PORCUPINE study will recruit c 40 RNF43 and/or RSPO fusion selected patients with aberrant microsatellite stable (MSS) metastatic colorectal cancer (mCRC) that have progressed following current standard of care treatment. This follows an open label, multi-centre, multi-arm format and includes a monotherapy arm (recruitment underway) and a combination arm with nivolumab (due to begin enrolling in H122 COVID restrictions permitting). The dosing for the combination will be confirmed once the Phase I study has reported. Each arm will aim to have c 20 evaluable patients. Baseline and on-treatment biopsies, as well as ctDNA testing and FDG PET scans, will be used to assess changes to tumour and immune micro-environment. All patients will have prophylaxis with denosumab. Initial results could become available during H123.

Redx has an open IND in the US which permits recruitment of US patients for the first time. Redx’s strategic partnership with Caris Life Sciences should help accelerate US recruitment of MSS mCRC patients into the PORCUPINE trial. Caris’s molecular profiling and screening capabilities enable the identification of eligible patients more quickly, that can then be trial matched with its own network of US trial sites.

The PORCUPINE2 monotherapy trial is an open label, multi-centre, multi-arm, study to evaluate efficacy and safety of a 2mg once daily dose of RXC004 in c 30 patients with Wnt ligand driven tumours. The first arm will recruit patients with RNF43 loss of function (LoF) mutation-positive pancreatic ductal adenocarcinoma (PDAC). The second arm will recruit patients with biliary tract cancer (BTC), where selection will not be required as BTC typically has a high Wnt ligand drive (>70% of cases). Again, all patients will receive denosumab prophylaxis. The aim is to have c 15 evaluable patients enrolled in each arm. These tumour types have limited treatment options and poor five-year survival rates (<3% for biliary tract and pancreatic cancer).

Four other Porcupine inhibitors in the clinic

In addition to RXC004, four other Porcupine inhibitors are known to be in clinical development for oncology indications:

• WNT974 (previously LGK974, Novartis): the lead player for some time, with WNT974 having successfully completed Phase I monotherapy trials in various solid tumours. A 27 patient Phase I study in combination with Novartis’ development-stage PD-1 checkpoint inhibitor spartalizumab showed promising signs of efficacy and a manageable side-effect profile. The most recent Phase I study was also supportive, showing WNT974 was well tolerated, with biomarker analysis highlighting possible enhancement of CPI activity.
• ETC-159 (A*STAR): a small molecule in development by Singapore’s Agency for Science, Technology and Research. It is listed (December 2021) as undergoing Phase I dose escalation trials both as a single agent and in combination with pembrolizumab in Singapore and the US in advanced solid tumours. Interestingly, the long half-life requires dosing on alternate days. There is little in the public domain.
• CGX1321 (Curegenix): CGX1321 is the lead programme for this US/China (San Francisco and Guangzhou) based clinical-stage company. Again while there is little information in the public domain, CGX1321 has successfully completed a single-agent dose-escalation Phase I study programme, with a Phase Ia single agent China study in gastric tumours and a Phase Ib US trial in combination with pembrolizumab in CRC still underway.
• XNW7201 (Sinovent): Another lead programme which is in Phase I trials in solid tumours in Australia exploring PK, safety, tolerability and seeking to establish a maximum dose. Suzhou, China based Sinovent was founded in 2017 and has three clinical programmes with five at earlier stages of development. A combination study with Alphamab Oncology’s KN046 (a bispecific PD-L1/CTLA-4 antibody) was announced in June 2020. Again, there is limited public domain information.

The WNT974 Phase I study (Rodon), published in July 2021, provides useful context for RXC004’s monotherapy data at ESMO. The data showed good tolerability, with the most reported adverse events of dysgeusia (loss of taste), gastrointestinal events, and fatigue. Dysgeusia (47% of patients) and bone-related disorders (6%) are consistent with Wnt pathway inhibition and can be considered on-target effects. Although the study evaluated 98 patients, these were not screened for the appropriate genetic mutations. No responses were seen by RECIST criteria; 16% of patients had stable disease (median duration 19.9 weeks). Examining individual patient data suggests 28 patients had an appropriate genetic profile (loss of function RNF43 or RPSO fusion without downstream APC mutations) and the anti-tumour activity was found here. Clearly, while encouraging, such retrospective analysis should be considered as indicative only.

Main takeaways from the R&D day KOL presentations

Redx’s October 2021 R&D Day provided additional detail on both RXC004 and RXC007, with acknowledged experts (Key Opinion Leaders, KOLs) adding valuable insights from a clinical and regulatory perspective. The KOL for Porcupine inhibition was Professor Scott Kopetz, Department of Gastrointestinal Medical Oncology, The University of Texas MD Anderson Cancer Center, and his presentation addressed the nature and size of the problem with mCRC.

CRC is the third most common cancer in US and globally accounts for 2m new cases annually and c 1m deaths. If metastasised (22% of patients present with mCRC at diagnosis and c 70% of earlier stage CRC progress to mCRC) outcomes are poor, with median survival of 30 months and five-year survival of 14%. Once the first two standard treatment regimens (oxaliplatin- or irinotecan-based combinations with 5-FU or capecitabine and then VEGF and/or EGFR inhibitors), have been exhausted the median PFS and OS drop to c 2m and 6m respectively.

Microsatellite stable (MSS) disease accounts of c 95% of all mCRC. Much research effort seeks to improve outcomes for genetically selected sub-populations. RXC004 is addressing one of the larger sub-populations (c 8% are driven by RNF43m or RSPO fusion LoF mutations) as monotherapy and in combination with a CPI. Porcupine inhibition is anti-proliferative rather than cytotoxic, which means a durable disease stabilisation is viewed as a clinically meaningful outcome. Regulatory agencies are increasingly comfortable with use of PFS as surrogate for OS in CRC. Prof Kopetz confirmed the FDA would, in his view, be comfortable using such data to support approval.

The key single message from these opinion leader discussions is that careful patient selection for the appropriate tumour genetic profile is an essential prerequisite for any Porcupine inhibitor monotherapy and, assuming the proposed mechanism of action is viable, the planned Phase II trial should be able to demonstrate statistically significant clinical efficacy. Aside from any anti-proliferative effects, the value of Porcupine in preventing tumour immune evasion could allow checkpoint inhibitors to be employed in these settings. Success here would not simply address a genetically specified tumour population but could help unlock the potential of CPIs in mCRC.

RXC007: approaching Phase II initiation

RXC007 is a novel and highly specific small molecule that selectively targets the ROCK2 (Rho Associated Coiled-Coil Containing Protein Kinase 2) receptor. It embarked on its first Phase I study in healthy volunteers in June 2021. Interim results were presented at the October 2021 R&D day and showed RXC007 to have a promising safety profile. The study is continuing dose escalation and is expected to read out in H122. These data will guide the dosing and structure of the RXC007 Phase II programme, expected to start in 2022.

Exhibit 9: ROCK is a central node in fibrosis-associated signalling pathways

Source: Rocking the Boat The Decisive Roles of Rho Kinases, Saadeldin et al, Front Cell Dev Biol Jan21

ROCK is a biologically and clinically validated target that has been shown to sit at a nodal point in a cell signalling pathway, where it modulates inflammatory response and fibrotic processes. The ROCK pathways mediate a broad range of cellular responses that involve the actin cytoskeleton and are important regulators of cellular growth, migration, metabolism, and apoptosis (Exhibit 9). The importance of ROCK has been known for some time, but the chemistry is particularly complex and historically identifying safe and effective selective inhibitors has proved challenging.

There are two kinase forms, ROCK1 and ROCK2, that share an overall 65% homology in amino-acid sequence and 92% homology in their kinase domains. They have broadly similar downstream functions, and both play critical roles in many cellular responses to injury (especially in fibrosis). For instance, the pharmacological inhibition of both ROCK isoforms has been shown to prevent airway remodelling and lung fibrosis. However, the simultaneous targeting of both forms appears to be more closely associated with cardiovascular effects (notably hypotension), hence the clinical importance of selectively targeting ROCK2 for a systemic agent.

Fibrosis is a critical factor in many debilitating diseases

Fibrosis occurs when the normal healing process goes awry, with the formation of excessive scarring. It develops because of aberrant wound healing responses to repetitive injury. Tissue responses to injury involve coordinated activities of multiple cell types that, when appropriate in duration and magnitude, restore normal tissue structure and function. Fibrosis can occur in most tissues, and severely impairs the function of the affected organ. The initial causes of fibrosis are manifold, and while the precise disease process is not fully understood. It typically involves a common series of events, including secretion of cytokines which provoke a pro-fibrotic, chronic inflammatory immune response that leads to production of excessive extracellular matrix (ECM) proteins (eg collagen) and the tissue becoming fibrous in nature.

Exhibit 10: ROCK’s role in fibrosis signalling and remodelling

Source: Matrix Stiffness the Conductor of Organ Fibrosis Santos 2018

Pro-fibrotic signals are delivered to cells after injury by both biochemical mediators and mechanical forces, and ROCK activation is central to many cellular responses to both types of signals (Exhibit 10). LPA (lysophosphatidic acid), thrombin, and TGF-β (Transforming Growth Factor-β) are important mediators that appear to act through ROCK and inhibition of the ROCK receptors can block the pro-fibrotic progression. There is mounting evidence that the behaviours of the cells involved in these wound healing responses, particularly epithelial cells, endothelial cells, and fibroblasts, are fundamentally regulated by ROCK signalling. ROCK activation has been implicated in the development of fibrosis in multiple organs including the lungs, heart, liver, kidneys, peritoneum, and skin.

RXC007 has particularly encouraging preclinical data

Redx explored a series of highly selective and orally active ROCK2 inhibitors in in vivo preclinical models of multiple diseases with underlying fibrosis, including liver, lung, and kidney. Good ADME profiles and robust anti-fibrotic effects were seen and suggest a benefit in diseases such as idiopathic pulmonary fibrosis (IPF), non-alcoholic steatohepatitis (NASH), and diabetic nephropathy (DN). In H120 RXC007 was selected as the lead candidate and entered Phase I studies in June 2021 with a subsequent clinical trial plan in development. RXC007’s preclinical data package is very promising and suggests it has the potential to be a best-in-class compound. The ADME, toxicology profiles (notably with liver enzymes), and on-receptor activity are well suited for fibrosis indications. The clinical programme is directed, at least initially, towards the more serious indications, such as IPF, which have significant mortality and where the medical need is greatest.

Exhibit 11: Significant market opportunity in IPF

Source: Redx Pharma  Note: (1) clinical estimates (Hyun 2015, Ley 2012, Rahu 2006); (2) patient numbers (diagnosed prevalence) and market size forecast (GlobalData: US, EU5, Japan)

The breadth and range of potential ROCK2 indications are many, extending from ophthalmic to intestinal. Aberrant downstream signalling is known to have important roles in cardiovascular diseases, CNS disorders (including Alzheimer’s and Parkinson’s), as well as diabetes (including insulin resistance and nephropathy) and a range of fibrotic dysfunctions. The main fibrotic indications alone are sizeable and poorly addressed by currently available therapies:

• Idiopathic pulmonary fibrosis (IPF) involves irreversible scarring of the interstitial cells in the lungs. There are no effective treatment, although drugs such as pirfenidone and nintedanib have shown slowing of disease progression. Palliative care includes the use of steroids, oxygen therapy, and pulmonary rehabilitation. Lung transplants are often the only resort. Life expectancy from diagnosis is typically three to five years, in line with that seen with aggressive cancers. Over 50,000 new cases are seen per annum in the seven largest markets. Estimates of market size vary between $2bn-$3bn pa but fail to include the non-pharmaceutical treatment costs and impact on quality of life.
• Non-alcoholic steatohepatitis (NASH) is a progressive inflammatory and fibrotic disease of the liver. It sits on a pathway starting with non-alcoholic fatty liver disease (NAFLD) and often ends with hepatocellular carcinoma (HCC). It is linked with obesity, diabetes, and sedentary lifestyles. Incidence is rising globally, and the current prevalence of 1.5% to 6.5% is expected to rise significantly. There are no approved products as yet, but promising approaches range from Vitamin E to liraglutide (GLP-1 agonist). Addressing the fibrotic challenge remains the pressing need. Estimates of likely market size range from $18bn to $37bn.
• Diabetic nephropathy is a complication of diabetes that will affect between 20% and 40% of all patients. It is the leading cause of chronic kidney disease and end-stage renal failure. Given 8-10% of the global population is expected to develop diabetes, the magnitude of the issue will become material. Tight blood glucose control and the lowering of hypertension are known to delay progression, as have angiotensin II receptor blockers (ARBs) and angiotensin-converting enzyme (ACE) inhibitors. Newer therapies such as SGLT2 inhibitors and GLP1 agonists are beginning to make a difference too. However, there remains a clear need for novel treatments that address the structural changes that are happening within the kidney.

RXC007’s development will initially be focussed on IPF, which has been selected as the lead indication because of the clear clinical need, limited treatment options, and sizeable commercial opportunity (Globaldata estimates the IPF market will be worth $3.6bn by 2029).

RXC007 completing Phase I studies in H122

Interim data from the Phase I study, in healthy volunteers, showed an excellent safety and pharmacokinetic (PK) profile. The dose escalation started with a single dose of 2mg and has risen through to 40mg. No adverse effects have been seen and the PK curves show a linear exposure, confirming the predictions from the preclinical studies. Although already at the predicted biologically relevant doses, further escalations are continuing. At the 40mg dose the half-life was 11 hours, confirming that the hoped for once daily dosing is viable. Full results from this study are expected in H122 and these will guide the dose for Phase II.

Exhibit 12: RXC007 Phase II programme design in IPF

Source: Redx Pharma  Note: Pb = placebo; FVC = forced vital capacity; DLCO = carbon monoxide diffusion coefficient; HRCT = high resolution computerised tomography

The staged Phase II study in IPF is expected to start in 2022 (Exhibit 12). This consists of a Phase IIa safety and tolerability study with and without standard of care (SoC). Currently there are two approved products (pirfenidone and nintedanib), but their limited utility and side-effects mean SoC treatment regimens vary. Indeed, many development candidates in IPF have failed due to tolerability issues. The RXC007 study will use biomarkers to help understand target engagement as well as providing early efficacy readouts, with these results guiding the dose for the Phase IIb portion of the trial. The Phase IIb will dose RXC007 over 12 months plus SoC with FVC lung function as a primary endpoint.

Kadmon acquired by Sanofi for $1.9bn

Few selective ROCK2 inhibitors are known to be in clinical development: RXC007 is believed to be one of only three. The ROCK class was recently the focus of industry attention following Sanofi’s all-cash acquisition of Kadmon for $1.9bn (a 79% premium) in September 2021. The deal centred on Rezurock (belumosudil), a selective ROCK2 inhibitor that is a first-in-class treatment for chronic graft-versus-host disease (cGVHD) for use in adult and paediatric patients 12 years and older who have failed at least two prior lines of systemic therapy. The July 2021 FDA approval of Rezurock established a clear regulatory pathway that, in our view, materially de-risks RXC007’s development process.

Kadmon was also developing a second ROCK inhibitor, KD045, that has generated positive preclinical results in lung, kidney, and liver fibrosis models. KD045 is a pan-ROCK inhibitor, opening concerns about possible hypotension due to inhibiting ROCK1 and ROCK2 simultaneously (see earlier). The only other clinical stage selective ROCK2 inhibitor we have been able to identify is TDI01, a Phase I asset licenced by Graviton Bioscience from Sino Biopharmaceutical subsidiary Beijing Tide Pharmaceutical in February 2021.

AZN5055: AstraZeneca starts clinical studies

AZN5055, previously known as RXC006, is a potent small molecule of the Porcupine receptor in development by AstraZeneca for fibrosis indications. As mentioned earlier in the oncology discussion, the Wnt pathways are critical elements in maintaining adult cell homeostasis, which includes wound healing and repair functions. Here aberrant wound healing causes increased proliferation and attenuated apoptosis of myofibroblasts, which results in the excessive synthesis, remodelling, and contraction of extracellular matrix that characterises fibrosis. Myofibroblasts are the key cells in the pathophysiology of fibrotic disorders and their differentiation can be triggered by multiple stimuli, with Wnt being one of the three key elements (the others are TGF-β and YAP/TAZ signalling). As in the oncology indications, Porcupine inhibition may play a useful role here.

We note that AZN5055 belongs to a different chemical class to RXC004, with an independent family of patents. It has shown promising efficacy and tolerability in a number of preclinical fibrosis models (including lung, liver and kidney), with a poster presented at European Respiratory Society (ERS) 2019. The porcupine route also represents a different approach to fibrosis to the ROCK pathways that are being progressed in-house with RXC007.

AZN5055 was licensed to AstraZeneca in August 2020 in a deal consisting of an upfront fee and early development milestones totalling $17m, further development milestones worth up to $360m, and mid-single digit royalties on any eventual sales. Several milestones have since been achieved, with a $4m milestone received in June 2021 on the progress of AZN5055 towards the clinic, and a further $9m triggered in December 2021 on the initiation of first-in-human Phase I studies. Apart from the cost aspects of progressing another large clinical programme, the out-licensing of AZN5055 was a strategic move to lessen the exposure Redx had to the Porcupine class as a whole. Deal terms are attractive for a preclinical asset, retaining material financial upside if the clinical development progresses as hoped.

AZN5055 is the fourth compound originated by Redx’s discovery platform to successfully progress into clinical development. AZN5055’s clinical programme is starting with a two-part Phase I study involving 104 patients. The primary indication appears to be IPF, which, if these early studies are supportive, we expect will be broadened into other fibrosis indications.

Jazz Pharmaceuticals: a partnership working well

Redx Pharma is working with Jazz Pharmaceuticals to develop a pan-RAF inhibitor programme (known as JPZ815) for RAS and RAF mutant tumours. The aim is to overcome the resistance mechanisms that are seen with clinically approved B-RAF selective drugs. The RAF kinases are an integral part of the RAS-RAF-MAPK pathway, with B-RAF mutations commonly seen in the clinic.

This deal was agreed in July 2019, with Jazz paying an upfront fee of $3.5m and Redx eligible for up to a further $203m in success-based development, regulatory, and commercial milestones, and mid-single digit royalties on eventual sales. A $3m milestone was triggered in September 2021 when IND-enabling studies were initiated, with the next milestone due on IND submission to the FDA is made (potentially in 2022). Jazz is funding all development work. Redx has a separate collaboration agreement, signed in parallel, to perform research and preclinical development services to completion of IND-enabling studies.

The MAPK (mitogen-activated protein kinase) pathway plays a critical role in the proliferation of numerous cancers, being seen as a growth driver in over a third of all solid tumours. The main downstream cascade involves the RAS protein family, and, in turn, the RAF kinase groups. There are three RAF enzyme members – A-RAF, B-RAF, and C-RAF – with B-RAF currently seen as particularly clinically relevant as it is mutated in 50-70% of malignant melanomas, 40% of thyroid carcinomas, 30% of ovarian tumours, and nearly 100% of hairy cell leukaemias. Within this, the V600 mutation is the most prevalent and active, so became a key target and led to the introduction of initially highly effective first-generation compounds, such as vemurafenib and dabrafenib.

However, not only did treatment resistance surface quickly but, in what has become known as the RAF inhibitor paradox (or RAF dimer dilemma), the drugs activated the MAPK pathway elsewhere. These targeted first-generation products actively triggered compensatory feedback loops in tumour cells and in the components of the tumour microenvironment. A number of approaches are being explored, including the use of pan-RAF inhibitors, with promising efficacy as monotherapy in animal and preclinical models. Several pan-RAF inhibitors are in Phase I development, including LXH254 (Novartis), TAK-580 (Takeda/Sunesis), which appears to have a poor profile, and HM95573 (Genentech/Hamni). LY3009120 (Eli Lilly) was too toxic at therapeutic doses, hence failed to show a benefit and was terminated.

Redx’s relationship with Jazz Pharmaceuticals was reinforced through a further research collaboration (September 2020) to discover and develop drug candidates for two oncology targets on the Ras/Raf/MAPK pathway. Deal economics include a $10m upfront payment and a further $10m in the following 12 months (triggered in December 2021 reflecting continued progress), with up to a further $400m in development, commercial, and regulatory milestones split equally between the two programmes, and tiered mid-single digit royalties on net sales. Redx is responsible for research and preclinical development up to IND submission, which will trigger the first milestone. After this point, Jazz is responsible for funding and conducting development.

Discovery efforts still moving at pace

Redx Pharma’s discovery engine is the source of both its in-house assets and those that have been sold/partnered in lucrative deals given the stage of development. Revenue from partners generated to date has been a useful source of funding and has provided important scientific validation. The partnerships with AstraZeneca and Jazz Pharmaceuticals (and sale to Loxo Oncology) were struck at the research/preclinical phase and the former are associated with attractive economic terms, which include potential success-based milestones of $1bn in aggregate plus royalties on sales.

Management aims to have three further wholly owned reaching IND submission by 2025. The research focus remains on the attractive pathways that impact selected oncology and fibrosis indications. Aside from the known, and sizeable, clinical disease areas, there is a growing industry interest in how fibrosis may be implicated in tumour resistance to treatment and in cancer associated fibroblasts. For instance, various fibrosis pathways have been associated with the stabilisation of a tumour extracellular matrix. These can act as effective functional barriers against agents, including immuno-oncology therapies, and their disruption could materially improve treatment outcomes both directly and as part of a combination approach. These pathways are known to be complex and difficult to model, hence present opportune themes for deploying Redx’s medical chemistry expertise.

Exhibit 13: Validated discovery engine fuels pipeline

Source: Redx Pharma   Note: ADME = absorption, distribution, metabolism, excretion; IND = Investigational New Drug Application; PoC = proof-of-concept; PK = Pharmacokinetic

Pan-ROCK GI-acting inhibitor the most advanced

The discovery programmes are known to include a second in-house ROCK asset, referred to as GI-targeted ROCK. This is at the research stage but has shown interesting preliminary data that lends itself to addressing inflammation and fibrosis of the gastrointestinal tract.

The gastrointestinal tract has a remarkable ability for self-regeneration following short-lived and mild insults, as in peptic ulceration, infectious enteritis, or mild diverticulitis. However, if inflammation becomes chronic and severe, as in Crohn’s disease, inflammatory mechanisms drive the excessive production of extracellular matrix (ECM) components and activate intestinal stromal cells that produce fibrosis. Even in the absence of inflammation, tissue damage and fibrosis continue to progress with increased accumulation and crosslinking of ECM.

Once fibrosis is established, control of inflammation with even biologics is not sufficient to halt fibrosis progression as matrix stiffness can drive fibrosis independently of intestinal inflammatory activity. Hence anti-inflammatory treatment is best suited for early-stage disease, as fibrosis might become self-perpetuating once ECM activity has become established.

In ulcerative colitis the fibrosis is located mainly in rectal mucosa and submucosa, but in Crohn’s disease it can be seen in all regions of the intestinal wall. Between a third and a half of Crohn’s disease patients develop clinically relevant fibrostenosis that leads to hospitalisations and endoscopic interventions or surgery. Between 70-90% of patients will require at least one surgical resection within their lifetime, with a recurrence rate of up to 70%. The patient, and economic, burden of fibrotic strictures in Crohn’s disease is significant.

The ROCK receptors are expressed in fibroblastic, epithelial, endothelial, and muscle cells of the human intestinal tract and are activated in inflamed and fibrotic tissue. Redx Pharma has evaluated pan-ROCK inhibitors that address both ROCK1 and ROCK2 pathways in several preclinical and animal models. These have shown inhibition prevented myofibroblast accumulation, expression of pro-fibrotic factors, and accumulation of fibrotic tissue; repeated administration resulted in the prevention and reversal of the fibrotic damage. The results suggest ROCK inhibition produces both prophylactic (prevents) and therapeutic (reverses) effects, which should translate into potentially clinically meaningful benefits.

Redx Pharma has taken a very interesting, and innovative, approach. The pan-ROCK inhibitor selected is designed to only work locally in the gut wall and, as it is quickly degraded by metabolic enzymes, to have a short half-life once absorbed. The aim is to avoid the systemic side-effects, notably cardiovascular (smooth muscle relaxation of systemic vasculature leading to hypotension), that are associated with simultaneous inhibition of ROCK1 and ROCK2. A poster of the preclinical data for REDX08087 was presented at ECCO (European Crohn’s and Colitis Organisation) 2018. Preclinical work to select a lead compound is underway and a development candidate is expected to be chosen in 2022. The novelty of the approach will likely need the clinical programme, and endpoints, to be discussed with the regulatory agencies.

DDR inhibitors in evaluation for fibrosis indications

Management has also disclosed that it has a research-stage DDR (discoidin domain receptor) inhibitor programme, with lead optimisation the next expected milestone. There are two discoidin domain receptors, DDR1 and DDR2, which sit at the intersection of two important receptor families, the extracellular matrix (ECM) and tyrosine kinase receptors (TKI), that are actively involved in normal development and tissue homeostasis. Their role in important aspects of cell behaviour, including proliferation, migration, adhesion, and ECM remodelling, has recently brought them to investor attention for solid tumour indications. For example, in November 2021 Parthenon Therapeutics raised a $65m Series A round to fund its preclinical research into reprogramming the tumour microenvironment and progressing its lead asset, DDR1 inhibitor PRTH-101.

DDRs differ from other kinase enzymes in acting on collagen, both fibrillar and non-fibrillar, and are the only TKIs that are directly activated upon contact of cells with their collagenous matrix. Both receptors recognise collagen as their ligands but DDR1 will bind and signal in response to both fibrillar and non-fibrillar collagens, whereas DDR2 is mostly activated by fibrillar collagens. The role of these receptors is still being explored, but DDR1 is expressed widely in developing and adult tissues, particularly in the epithelia of skin, lung, liver, kidney, gut, colon, and brain, whereas DDR2 shows higher expression in mesenchymal tissues such as heart, muscle, and connective tissues.

Once bound to collagen, DDRs undergo autophosphorylation (which is unusually slow, taking hours rather than the more typical seconds) and initiate several signalling pathways involving the ECM and control adhesion and cell motility. The interactions are complex and varied, regulating key cellular processes including protease production, cytokine secretion, cell migration, immune cell recruitment, and matrix production. In fibrosis, DDR expression, mainly DDR1, is up regulated and results in initiation and progression of fibrosis (notably in lung and kidney). DDR mediated signalling also appears to have an important role in several fast-growing invasive tumours, as key drivers of proliferation and stromal invasion.

Redx has developed a number of potent and selective DDR inhibitors that are undergoing lead optimisation with preclinical studies ongoing. These are initially being explored in models of fibrotic disease and offer the potential of disease modifying activities. Currently there is little public information available but, once again, management has demonstrated its ability to successfully address difficult and complex pathways with attractive small molecules.

Sensitivities

Typically, with innovative healthcare companies the three main sensitivities relate to the clinical and regulatory aspects, commercial execution, and the financial resources required to accomplish these. More specifically for Redx, the key near- and medium-term sensitivities are directed to the clinical progress with the four lead programmes, two in-house and two partnered.

Management’s strategy is to focus on small molecules that are either first-in-class or best-in-class. Addressing highly novel targets clearly carries a greater risk, however the scientific validity and clinical relevance of the mechanism has been elucidated. Importantly, the novelty of the compound, while making the development process more challenging, means that it is likely to be commercially attractive to prospective partners. Creating a best-in-class molecule carries less risk as learnings from the leading players are applied. However, the timeliness of the development becomes paramount as speed of development is a critical factor.

The risks of clinical development are well known and documented. Less than 8% of preclinical programmes ultimately reach the market. The success probabilities improve as a programme progresses through clinical development, with a key inflection point seen at the Phase II proof-of-concept stage. This is often viewed as an attractive point for value optimisation as the risk profile improves materially but the more expensive, and pivotal, Phase III trials lie ahead.

The partnering process is the key test of a management’s strategy. A well-struck deal validates not simply the attractiveness of the proprietary technology and scientific skills, but the commercial terms are a tangible insight into management acumen. Given its size and history, Redx has an impressive track record of developing commercially attractive targets. Three preclinical assets have been successfully out-licensed, which suggests the inherent value of these individual programmes is such that it recognised ahead of clinical proof-of-concept studies.

Financing is a perennial element to any innovative research-based company and Redx is no exception. We believe the strategy to develop selected assets to a greater value creation point is sound, the inherent scientific expertise is proven, and the current management is well respected. The real question is whether investors can appreciate the investment case and support Redx Pharma through to the next phase of its journey. The presence of industry specialists such as Redmile, Sofinnova, and Polar Capital on the register is reassuring and the past opportunist approach by Samuel Waksal (via Yesod BioSciences) suggests that canny industry players can see the inherent value. This background suggests that funding will become available when the need arises.

Valuation

Redx Pharma is a classic discovery and development play, hence is well suited to valuation using an rNPV model. However, such models tend to attribute most value to later stage clinical compounds and underplay earlier stage programmes. To counter this a value must be included for the discovery platform and, understandably, this requires more subjective considerations than a simple rNPV calculation. In Redx’s case the track record of generating attractive compounds and management history of striking commercially sound licensing deals gives a satisfying degree of comfort that our valuation remains realistic but, in line with our philosophy, still errs on the side of caution.

The rNPV of the individual development projects are assessed and success probabilities adjusted for the inherent clinical, commercial, and execution risks each carry. These are summed and netted against operational costs and net cash. We also include risk-adjusted development milestones for actual and assumed licensing deals, which are benchmarked against similar deals. The success probabilities are based on standard industry criteria for the respective stage of clinical development but, importantly, flexed to reflect the inherent risks of the individual programme, the indication targeted, and the trial design.

Even though the strategy envisages the out-licensing of at least some of the programmes before the later, and more expensive, stages of clinical development, we allow for the commercial and execution risks as we view these as integral to any programme’s intrinsic value. As always, we employ conservative assumptions throughout our modelling, particularly regarding market sizes and growth rates, net pricing, adoption curves, and peak market penetration.

For the discovery platform we examined the historical output, particularly its quality and commercial attractiveness, and assessed the likely sustainable future output. As mentioned, the track record is impressive; the BTK programme that was sold to Loxo (sadly as a distressed sale) has progressed well and provides tangible evidence of Redx’s ability to produce high-value assets. Similarly, the AstraZeneca and Jazz Pharmaceuticals deals provide reassurance that Redx’s output is desirable and reproducible. The aim of generating an average of one lead drug candidate per annum may appear ambitious but, when placed into historical context, is realistic and achievable. It is against this framework that we attribute a value of between $160m (£123m) and $240m (£185m) for the discovery engine. Again, being conservative, we have opted to use £129m in our modelling. At present, the Jazz Pharmaceuticals Ras/Raf/MAPK collaboration is included within the discovery platform, although we intend to break this out with an explicit value once there is more clarity on timelines and indication(s).

Our model ascribes a Redx valuation of £433.6m ($564m), equivalent to 157.5p per share (105.1p fully diluted). Outputs and underlying assumptions of our model are shown in Exhibit 14. Looking at the elements of our valuation in greater detail:

• RXC004 is the most advanced programme and its valuation of £122.4m (44.5p/share) reflects its Phase II clinical stage and likelihood of successful progression. We have not modelled its use in an explicit oncology patient population, an exercise we would normally perform in the later stages of development, but we do highlight that development in relevant genetically defined cancers is pursued. However, even a cursory exploration of its potential use in these selected cancers, which could support pursuit of accelerated regulatory approval pathways and attractive pricing, suggests that our assumptions are justifiable.
• RXC007, valued at £87.8m (31.9p/share) is a material contributor to our model despite being deliberately cautious; we flag that this programme could surprise on the upside. Fibrosis is an attractive area due to the poor treatment options currently available and the number of debilitating conditions where it is a key element. We expect development to focus initially on more severe indications, such as IPF, but if efficacy and, importantly, tolerability is appropriate then larger indications such as NASH and diabetic nephropathy could become accessible.
• The AZD5055 valuation of £39.7m (14.4p/share) is based on risk-adjusted estimates for the development, registration, and commercial milestones from AstraZeneca together with an assumed 5% sales royalty. We assume typical timelines for development progression and approval.
• The Jazz Pharmaceuticals Pan-RAF asset JZP815 is treated similarly, with milestones for the usual inflection points and 5% royalties on eventual sales. The value is £22.4m or 10.2p/share.

Exhibit 14: rNPV-based valuation of Redx Pharma

Source: Trinity Delta   Note: The rNPV of RXC004 and RXC007 includes a deal success factor of 80%, and of 75% for GI-targeted ROCK; other valuation assumptions include a 12.5% discount factor, £/$ FX rate of 1.30, and 10% taxation from 2028 (UK patent box).

As mentioned, our conservative approach yields a valuation of £433.6m ($564m), equivalent to 157.5p per share. The clinical progress of the various pipeline assets should unlock upside, as further data would prompt us to adjust the respective success probabilities that reflect the inherent clinical, commercial, and execution risks that each programme carries. Additionally, as these programmes progress, there should be more insight into the specific oncology or fibrosis patient populations that will be addressed, and this in turn would mean that peak sales (pricing, penetration) and timeline assumptions could be revisited.

To provide context we have collated data from peers (Exhibit 15) with similar business models and a comparable small molecule R&D pipeline in terms of disease focus, size, and maturity. All are publicly listed and, except for Inventiva, are US based. Their stock market valuations range from c$130m to $2.4bn.

Exhibit 15: Small molecule discovery and development peers

Source: Trinity Delta, Company websites    Note: pricing as of close of business February 15, 2022

Looking at related corporate activity, Loxo Oncology was acquired in 2019 by Eli Lilly for $8bn (Loxo had acquired Redx’s RXC005, BTK inhibitor, for $40m in 2017) and Sanofi acquired Kadmon, for the ROCK2 inhibitor Rezurock (belumosudil), for $1.9bn in 2021.

Financials

Redx’s financials are relatively straightforward. Revenues consist of modest research collaboration income and more sizeable, albeit periodic, milestone receipts from partners; expenditure is largely investment in both discovery and in-house development activities, which will rise as programmes progress through the costlier clinical phases. Exhibit 16 outlines our forecasts and financial summary.

2021 was a year of delivery for Redx, with development progress of both in-house and partnered programmes. At end-FY21 (30 September 2021) the company’s cash position was £29.6m (end-FY20: £27.5m) having been boosted by the £25.7m (gross) proceeds from the Placing and Open Offer in December 2020, plus milestone receipts of £2.8m ($4m) and £2.2m ($3m) received during the period from partners AstraZeneca and Jazz Pharmaceuticals. Post-period end, a further £7.3m ($10m) was received from Jazz Pharmaceuticals and £6.5m ($9m) from AstraZeneca as key milestones were achieved. The level of investment in RXC004 and RXC007, as they continue into and through their Phase II trials, means current cash provides a runway through Q4 calendar year 2022 (Q1 FY23).

FY21 revenue of £10.0m (FY20: £5.7m) consisted mainly of milestone receipts (£5m), collaboration revenues (£2.75m) and payments for research and preclinical development services (£2.28m). Milestones from AstraZeneca are recognised on receipt (as they relate to contingent consideration on the license previously granted), whereas payments from the Jazz Pharmaceuticals collaborations (predominantly related to the underlying development services) have a deferred recognition element and are recognised as each stage is completed.

R&D expenditure was £24.4m (FY20: £10.5m), with the increase driven mainly by progress of the two in-house programmes (RXC004 and RXC007) through early clinical trials. R&D costs are expected to rise materially if the data supports continuing clinical development as each successive stage is costlier. Despite the increase in staff numbers and investment in support infrastructure, G&A expenses increased more modestly to £6.5m in FY21 (FY20: £4.2m). Finance costs were £1.7m (FY20; £1.0m). This translated into an FY21 net loss of £21.6m (FY20: loss of £9.2m), with loss per share of 8.4p (FY20: 5.4p). The cash outflow for the period was £21.4m vs a £0.4m inflow for FY20.

Looking ahead, we expect FY22 revenue of £12.0m. This is comprised of the $9m (£7.6m) AstraZeneca milestone receipt and recognition of £4.4m in revenues under the Jazz Pharmaceuticals Ras/Raf/MAPK collaboration (a similar amount is expected to be recognised in FY23). We do not currently include in our forecasts the anticipated milestone from Jazz Pharmaceuticals in connection with JZP815 IND submission with the FDA: this may potentially be triggered in 2022.

As indicated earlier, increased R&D investment will be the major driver of higher operating expenses in FY22 and FY23. We model R&D spend of £34.2m and £42.4m respectively.

Exhibit 16: Summary of financials

Source: Company, Trinity Delta     Note: Short-term debt in FY23E is indicative of our view of Redx’s funding requirement. Redmile/Sofinnova Convertible Loan Note has August 2023 conversion date, with a 15.5p conversion price, equating to a potential 110.3m of new shares. Revenue forecasts do not include any contribution from milestone payments yet to be received.

Contact details

Block 33
Mereside, Alderley Park
Alderley Edge
Macclesfield
SK10 4TG
United Kingdom

www.redxpharma.com

Top institutional shareholdings

Source: Redx Pharma   Note: Post CLN conversion, Redmile’s holding would be 69% and Sofinnova’s would rise to 15%.