Telix Pharmaceuticals — Targeted radiation to image and treat cancer

Telix is a biotechnology company that is developing molecularly targeted pharmaceutical products for imaging and treating a range of cancers. It is developing TLX250-CDx and TLX250 for imaging and treating the kidney cancer known as clear cell renal cell carcinoma (ccRCC); TLX591-CDx and TLX591 for imaging and treating prostate cancer; and TLX101-CDx and TLX101 for imaging and treating brain cancer (TLX101-CDx will be used for studying the pharmacology of TLX101 but will not be developed for clinical use). TLX250-CDx, which now uses the 89Zr radionuclide that produces sharper images, is in a bridging study ahead of the confirmatory ZIRCON Phase III trial expected to commence in Q418. The TLX101 therapeutic agent is expected to enter a Phase I study in the current quarter. A drug master file (DMF) for the TLX591 prostate cancer imaging kit has passed review by the US FDA. TLX591-CDx is now available for use in clinical trials, and a commercial launch as an investigational imaging tool is expected in Q318.

Our initial valuation of Telix is A$303m or A$1.39 per share, which includes our estimates of the future milestone payments and royalty streams for TLX250 and TLX591, plus profits from commercialisation of TLX250-CDx, TLX591-CDx and TLX101. Our rNPV calculation is based on the assumptions discussed below, such as target population, market uptake, pricing, R&D costs, market exclusivity expiry and calculated peak sales. Telix’s strategy is to self-commercialise TLX250-CDx, TLX591-CDx and TLX101 due to the compact market structure for these products. For TLX591 and TLX250 its strategy is to seek a marketing partner to promote the products into the more dispersed urology market. We assume the trigger for a licensing deal could be positive Phase I/II or Phase II data for the TLX250 and TLX591 therapeutic agents.

Telix reported an operating loss of A$6.4m in 2017, its first year of operation. Expenses associated with R&D projects were A$3.0m, personnel expenses were A$1.3m and administration and consulting costs totalled A$2.3m. Our total operating loss estimates for 2018, 2019 and 2020 grow to A$13.0m, A$17.9m and A$15.8m, respectively, mainly due to increased R&D expenditure as Telix progresses its clinical trial programme, partly offset by the Australian government’s R&D rebate scheme. Telix had A$42.0m cash and equivalents at 30 June 2018 and is funded into 2020. However, we estimate that it may need additional funding in 2020 (we model of A$4m of indicative debt) if it does not enter a licensing transaction before 2021.

Telix is subject to typical biotech company development risks, including the unpredictable outcome of trials, regulatory decisions, success of competitors, financing and commercial risks. Our model assumes that TLX250 and TLX591 will be out-licensed; therefore, our valuation is sensitive to potential licensing timing and actual deal terms. While Telix is commencing commercialisation of TLX591-CDx, it is still mainly a mid-stage drug developer, therefore in the foreseeable future most value creation will depend on successful R&D progress and any potential partnering activities. If the FDA requires Telix to conduct a prospective clinical study of TLX591-CDx rather than a blinded review of existing scans, it could delay filing for full marketing approval by one to two years. We model use of TLX591-CDx as a second-line test in prostate cancer patients following biochemical relapse when metastases cannot be detected by CT or MRI scans, but if it becomes a first-line test in these patients, the addressable market could be significantly larger.

Telix Pharmaceuticals is based in Melbourne, Australia and focuses on the development of MTR products (also referred to as targeted radiopharmaceuticals) to image and treat cancer. The company was incorporated in January 2017 and listed on the Australian Stock Exchange (ASX) in November 2017 after raising A$50m (before costs) in an IPO. The company’s founders, Christian Behrenbruch (CEO) and Andreas Kluge (CMO), are experienced executives and drug developers in the field of radiopharmaceuticals. The company’s portfolio consists of a pipeline of in-licensed and acquired products as well as company-originated intellectual property (Exhibit 1). The key transactions are summarised below.

In January 2017 the company in-licensed the rights to the underlying antibody technology behind the TLX250 programme from Wilex (Deutsche Börse ETR: WL6).

In July 2017 the company inlicensed from Abzena PLC (LON: ABZA) several patents that form the basis of the TLX591 portfolio. Telix is working with Abzena on a number of protein engineering enhancements to its programmes.

Telix has a collaborative agreement with the privately held French biotech Atlab Pharma SAS around clinical data and manufacturing process development for TLX591. Telix is likely to exercise its option to acquire Atlab for ~US$10m in cash or shares (we model 20.5m shares). Atlab licensed the huJ591 anti-PSMA antibody technology from Cornell and has rights to several patents that relate to the combination use of TLX591and other prostate cancer drugs including anti-androgen drugs.

In October 2017 Telix acquired the privately held German company Therapeia, obtaining the background technology to TLX101.

Exhibit 1 summarises the key features of the six products in the company’s portfolio.

MTR products selectively deliver radiation to cells with molecular markers or targets that are over-expressed in cancer. The radiation can be used to facilitate imaging a tumour to diagnose or stage a patient, or may be a therapeutic dose to treat the cancer. MTR products enable medical imaging, for example positron emission tomography (PET), to determine the true extent of disease and can also evaluate whether a treatment target is present. Exhibit 2 illustrates how MTR works.

The MTR products in use or in development use a range different of radioactive isotopes and targeting agents to address target molecular targets in cancer cells. For example, the most widely used PET imaging agent, 18F -fludeoxyglucose (FDG), which is widely used for detecting and monitoring cancer, relies on the fact that many tumours are metabolically active with a high glucose uptake. However, FDG PET is not very useful for imaging cancers such as prostate cancer that are not very metabolically active, or for organs where the signal from a tumour can be obscured by labelled FDG excreted in the urine.

Telix’s MTR imaging agents target molecular markers that are overexpressed on particular cancers. Their specificity for cancer cells makes them particularly useful for identifying small metastases that cannot be detected by other methods and for monitoring patients for recurrence or metastatic spread of the disease. In the case of TLX250-CDx, a key intended application is for distinguishing between benign lumps and dangerous kidney cancers.

Telix’s MTR therapeutics target delivery of radioactive payload to the tumour to kill cancer cells.

In September 2017 the EU approved an MTR therapeutic product known as 177Lu dotatate, or Lutathera, for the treatment of gastroenteropancreatic neuroendocrine tumours (GEP-NETs), a rare form of cancer. Lutathera was subsequently approved by the US FDA in January 2018. Lutathera reduced the risk of disease progression or death in NET patients by 79% in pivotal studies.

In October 2017, soon after the EU approval, Novartis announced an offer to acquire the developer of Lutathera, Advanced Accelerator Applications (AAA), for US$3.9bn.

In a particularly relevant example of pharma interest, in February 2014 Bayer acquired Algeta for 16.2bn Norwegian kroner (~US$2.6bn). Algeta had developed Xofigo, a therapeutic radiopharmaceutical for patients whose prostate cancer had metastasised to their bones, which had been approved in the US and Europe in 2013. Bayer had previously entered a licensing agreement with Algeta in relation to Xofigo in 2009.

In an unrelated development, in 2017 the FDA approved Xofigo (Bayer), a bone-targeted alpha particle called radium 223, for patients whose prostate cancer has metastasised to their bones.

The global radiopharmaceutical industry is expected to grow to US$26bn by 2030 (from US$4.5bn in 2016)1, which has stimulated increased interest from big pharma in the field. Telix is well placed to play an important part in the anticipated growth of the radiopharmaceutical industry.

Telix is developing TLX250-CDx as an imaging agent to detect ccRCC, which is the most common and aggressive form of kidney cancer; ccRCC comprises around 70-75% of all kidney cancers. TLX250-CDx uses a radiolabelled antibody that binds to carbonic anhydrase 9 (CA-IX), which is highly expressed on the surface of over 90% of ccRCC tumours, but is not detectable in normal kidney tissues. CA-IX is also highly expressed in a range of other cancers including lung, cervical, ovarian, oesophageal and breast carcinomas2.

TLX250-CDx is based on the mAb girentuximab (cG250) that Telix in-licensed from Wilex. Wilex developed an 124I-labelled version of girentuximab for imaging with PET, that it named Redectane. In the studies conducted by Wilex, PET/CT scans were done two to six days (average five days) after Redectane infusion.

Telix has improved on the Wilex imaging product by replacing the 124I radiolabel with 89Zr. 89Zr is a radio metal that becomes charge-trapped in the cytoplasm after the antibody bound to the CA-IX antigen on the surface is internalised and broken down within the cell, in contrast to 124I which is quickly released from the cell. The fact that 89Zr stays trapped within the cells results in an improved signal to noise ratio and clearer images, as can be seen in Exhibit 3. Exhibit 3 contrasts images obtained with the old 124I radiolabel on the left hand side with the much sharper images obtained with 89Zr-radiolabelled TLX250-CDx on the right.

ccRCC is the most dangerous kidney cancer and is the most likely to progress and metastasise (spread); it is four times as likely to develop distant metastases as the other less common, non-clear cell subtypes3. This means the key thing to know about a suspicious renal mass is whether or not it is ccRCC. This is becoming more important given that renal masses are increasingly being detected as incidental findings to abdominal imaging done for other reasons.

Unfortunately, in many cases current imaging techniques are unable to distinguish between ccRCC and benign tumours or less aggressive forms of kidney cancer. Surgical excision is the standard treatment of suspicious renal masses in patients who are otherwise healthy. However, a meta-analysis of 26 studies found that 15% of surgically resected renal masses were in fact benign4 (non-cancerous), so many patients are having unnecessary surgery.

At a meeting in 2012, the FDA’s Oncologic Drugs Advisory Committee (ODAC) concluded a test that could distinguish between ccRCC and other kidney masses would be useful when deciding how to manage renal masses. The ODAC concluded the test would be useful in the 20–30% of patients where the renal mass was considered indeterminate following examination by other imaging techniques such as ultrasound or CT. It was considered likely to be particularly useful for small lesions (1–2 cm) and for larger lesions in patients with poor overall health where the risks of surgery are greater.

Wilex conducted a US-based Phase III study of Redectane in 196 patients with renal masses that were scheduled for surgical resection. In 2010 the study reported sensitivity and specificity for ccRCC of 86% and 87% respectively5. While the sensitivity and specificity for Redectane were significantly higher than for the contrast-enhanced CT scans (CECT), the study missed one of its four primary endpoints because the specificity was not significantly higher than the pre-specified arbitrary target of 75% (p=0.057). This miss was due to the smaller than intended number of non-ccRCC patients in the study (53 vs the expected 63 non-ccRCC patients). Despite the 16/0 positive ODAC vote regarding the potential usefulness of an imaging test with the characteristics of Redectane, the FDA declined to approve Redectane and requested that Wilex undertake a confirmatory Phase III study; due to funding limitations Wilex did not conduct the requested study.

Having in-licensed the programme from Wilex, Telix has commenced the first stage of a confirmatory Phase III trial using its improved version of Redectane that it has termed TLX250-CDx. It has already recruited four of the 10 patients required for a dosimetry/bridging study to confirm the utility of replacing the old 124I isotope with the new 89Zr isotope, which is expected to complete in Q318. At the completion of the bridging study the ZIRCON Phase III trial will be expanded to recruit up to 200 patients at ~15 European sites (expected in Q418). Up to three Australian sites and up to eight US sites are also expected to join the study once it is up and running and to recruit up to 100 additional patients into the study. Management believes the Phase III study will complete by Q319.

According to the National Cancer Institute, kidney cancer is the eighth most common cancer, and is expected to account for 65,340 new cases and 14,970 deaths in the US in 2018. Globocan predicts that in 2020 there were will be 413,000 cases and 176,000 deaths from kidney cancer worldwide, while in the EU in that year there are predicted to be 94,900 new cases and 39,800 deaths.

At the time of diagnosis, the disease is localised to the kidney in 65% of cases, while it has spread to regional lymph nodes in 16% of cases and has metastasised elsewhere in the body in a further 16% of cases (staging is unknown in 3% of cases). While the overall five-year survival rate is 75%, the prognosis ranges from a five-year survival of 93% for patients diagnosed with localised tumours to a low 12% for patients who have metastatic disease at diagnosis.

Taking the midpoint of the assessment by the ODAC that an imaging test with properties similar to TLX250-CDx would be useful in 20–30% of patients with a renal mass, we assume that 25% of RCC patients would be candidates for an imaging indication similar to that sought by Wilex to assist in the characterisation of ccRCC as part of the diagnostic work-up of an indeterminate renal mass.

In addition to use in the initial diagnostic workup, the test could also be used to:

We suspect these additional indications could potentially be several times larger than that for initial diagnostic work-up of an indeterminate renal mass, but we conservatively model the potential market for these indications as the same size as that for the initial work-up (ie 25% of new cases), taking the total addressable market to 50% of new RCC cases each year.

High resolution CT is the first choice for imaging renal masses, but while a CT scan tells you about the structure of the mass it does not tell you about the biology. Currently, a renal mass that is enhanced on CT is considered malignant until proven otherwise as there is no alternative imaging agent that can identify ccRCC tumours.

A tissue biopsy can be used to identify cancerous renal masses, including ccRCC, but less than 10% of suspicious renal masses are biopsied in clinical practice due to the risks associated with collecting the biopsy tissue sample and because the rate of non-diagnostic biopsies is 10% to 20%6.

FDG-PET has limited utility for diagnosis of primary RCC as the renal uptake is likely to be masked by the excretion of the agent in the urine7; it is more useful for detecting metastases distant from the kidney.

An alternative strategy when investigating indeterminate real masses could be to actively highlight benign masses. A number of small studies have been conducted on the use of a mitochondrial imaging agent 99mTc-sestambi single photon emission CT (SPECT) to identify mitochondrial-rich benign and indolent (slow growing) renal masses such as oncocytomas and hybrid oncocytoma/chromophobe tumours. In a recent study8, 99mTc-sestamibi successfully identified all 7/7 ccRCC tumours as not being benign/indolent, and identified 16/24 (67%) of non-ccRCC tumours as being benign/indolent types.

As 99mTc sestamibi is already used for clinical investigations of myocardial perfusion and for the detection of parathyroid adenomas, it could potentially also be used to aid in the identification of benign/indolent renal masses in the clinic if sufficient evidence becomes available to justify its use. However, the lower image resolution of SPECT imaging means that it is not well suited to imaging small lesions. In addition, 99mTc sestamibi would not be suitable for any of the other potential applications of TLX250-CDx, such as detection of ccRCC metastases, assessing response to therapy or screening for suitability for TLX250 MTR therapy.

Telix is also developing TLX250, a MTR therapeutic targeting ccRCC. TLX250 is based on a modified and improved version of the mAb girentuximab (cG250) labelled with the isotope 177Lu. 177Lu is an ideal isotope for therapeutic MTR applications because it emits beta particles that travel only a short distance in tissue and are absorbed within the tumour, killing cancer cells. It has a comparatively short half-life of 6.64 days.

Wilex has completed two clinical studies of 177Lu-cG250 in patients with metastatic ccRCC. The two studies provided encouraging evidence of efficacy, with 78% and 68% of patients achieving responses of stable disease or better in the three months after the first course of treatment in the Phase I and Phase II studies, respectively. One of the 14 patients in the Phase II study achieved a partial response. While the treatment was generally well tolerated, a high proportion of subjects experienced myelosuppression (low levels of white blood cell production in the bone marrow). We expect the company to seek to optimise the dosing regimen (both amount and interval) in future studies to in order to reduce the incidence of this side effect. The company’s experience with MTR therapies in prostate cancer will help inform the programme to optimise the dosing regimen.

While many patients who are diagnosed with early-stage RCC can be cured by surgical treatment, better treatments are needed for metastatic disease, where the five-year survival rate is only 12%.

Checkpoint inhibitors and targeted therapies (antiangiogenic therapies and mTOR inhibitors) are the mainstays of systemic treatment for unresectable and metastatic ccRCC, rather than cytotoxic chemotherapy. According to clinical guidelines, the preferred first-line therapies are the targeted therapies pazopanib and sunitinib; ipilimumab + nivolumab combination (FDA approved April 2018) is a preferred first-line therapy for intermediate and low risk patients, but not for high-risk patients9.

Preferred second-line therapies are cabozantinib (TKI including VEGFR) and the anti-PD1 immune checkpoint inhibitor (ICI) nivolumab, which was approved by the FDA for patients with advanced RCC in November 2015.

Therapies such as ICI that enhance immune responses have had remarkable success in treating a range of cancers, however, typically less than a third of patients respond when these drugs are used as single agents (monotherapy). For example, the response to second-line treatment of mRCC with nivolumab was 22%10.

Combining ICI with chemotherapy has been an effective way to improve response rates compared to ICI monotherapy in a number of cancer types. For example, in newly diagnosed metastatic lung cancer the response rate to the Keytruda/chemo combo in the KEYNOTE-189 study was 49.4%11, which is substantially higher than the 27.3% ORR to Keytruda monotherapy reported in the separate KEYNOTE-042 study. The synergistic benefit of ICI/chemo combos is believed to be due to the cellular debris from tumour cells killed by the chemotherapy triggering an immune response (sometimes referred to as immunogenic cancer cell death or turning immunologically cold tumours hot), with the ICI then able to strengthen this immune response.

With chemotherapy not recommended in ccRCC, there is potential for MTR to play a similar role in killing tumour cells and releasing tumour antigens to stimulate an immune response. Combining MTR with ICI would be expected to improve ORR.

The ability of radiation therapy to have immune stimulatory effects is evidenced by the well-established phenomenon of so-called abscopal responses to radiation, where tumours away from the site of radiation also respond to treatment due to an immune response triggered by the therapy12.

Exhibit 4 shows how both chemotherapy and radiotherapy can cause immunogenic cancer cell death and thus play a role in initiating an anti-tumour immune response.

Furthermore, a randomised trial presented at ASCO in June provided direct evidence that radiation therapy can enhance responses to ICI therapy. In the PEMBRO-RT study in 64 patients with lung cancer, combining external beam radiation therapy with the ICI pembrolizumab improved ORR to 41% vs 19% for pembrolizumab alone. Median PFS was 6.4 months for the combo vs 1.8 months for pembrolizumab alone (HR 0.55, p=0.04)13.

We see good potential for TLX250 to be combined successfully with ICI therapy. However, it could potentially face competition from a form of external beam radiation therapy called stereotactic body radiation therapy (SBRT; also known as SABR) in patients who have only a small number of metastatic sites14.

Telix is formulating plans for three Phase II studies of TLX250 in combination with ICIs, one of which is likely to be a pharma-sponsored study. The three studies are expected to involve ~60 subjects in total. It plans to submit two INDs in relation to these studies in US/Australia in Q418. The manufacture of clinical grade material for these studies is expected to be completed by the end of August.

By the end of 2019 the company expects these studies to have produced sufficient information to identify an appropriate strategy for the Phase III development of TLX250. A key goal of the planned Phase II studies is to demonstrate that there are synergistic benefits from combining TLX250 with ICI therapy.

Telix is also developing TLX591-CDx (68Ga-HBED-PSMA-11), a small-molecule MTR diagnostic for imaging prostate cancer. TLX591-CDx targets the prostate specific membrane antigen (PSMA), which is expressed in very high levels in the prostate but in low levels on most other normal cells. PSMA is significantly over-expressed in prostate cancer cells and some other solid tumours15. Prostate cancer overexpression has been shown to be 100-fold to 1,000-fold that of normal tissue expression16.

TLX591-CDx is based on PSMA-11, a peptide ligand that binds to the PSMA protein. PSMA-11 labelled with the 68Ga radio isotope has been widely used in academic medicine in Europe and Australia to image prostate cancer, and has been shown to be safe to use and to produce high quality images (Exhibits 5 and 6).

While the PSMA-11 targeting agent (ligand) is in the public domain, the chemistry for linking the 68Ga radiolabel to the ligand and preparing the dose of the imaging agent with the TLX591-CDx kit utilises is proprietary chemistry developed by ANMI SA of Belgium.

ANMI has developed a so-called ‘shake and shoot’ kit. The kit contains all of the components needed to attach the 68Ga radiolabel to the PSMA targeting agent. The radioactive 68Ga is produced in the hospital pharmacy by elution from a column of 68Ge, which is gradually converted to 68Ga by radioactive decay. The kit takes only a few minutes to prepare an injection-ready dose, with all of the steps done at room temperature. The prepared radiolabelled product can then be injected into patient, who is then imaged in a PET scanner an hour or so later.

Telix has formed a JV with ANMI for the development and subsequent commercialisation of TLX591-CDx in the US. The US JV, known as Kyzeo Imaging, has completed validation at a US clinical site and a DMF has been reviewed and accepted by the US FDA. A DMF is a submission to the FDA that may be used to provide confidential detailed information about facilities, processes, or articles used in the manufacturing, processing, packaging, and storing of human drugs. A third party can incorporate the information in a DMF into an IND or NDA application by way of a letter from the manufacturer (Telix) to the applicant, which authorises reference to the DMF.

Now that the DMF has been reviewed, TLX591-CDx can be sold as an investigational product kit for use under an IND that references the DMF. The Memorial Sloan Kettering Cancer Center (MSKCC) recently launched an expanded-access study (NCT03204123), which allows the kit to be used to image prostate cancer patients at its clinics. Telix has granted Endocyte access to the DMF and is supplying TLX591-CDx kits to Endocyte for screening patients in its VISION Phase III prostate cancer trial.

Telix is in active discussions with distributors and plans to commercially launch TLX591-CDx in Q318. It has already encountered significant clinical interest or demand from over 50 hospital networks in the US. Telix believes its strategy of commercialising TLX591-CDx as an investigational test will give it the first 68Ga-based PSMA product for PET imaging widely available in the US. Imaging sites could be authorised to image patients using TLX591-CDx via a simple IND application supported by the DMF. As far as we could determine, no other company has submitted a DMF for a PSMA PET imaging product to the FDA17.

While TLX591-CDx is initially being commercialised as an investigational test in the US, Telix intends to seek full FDA approval, which would bring higher reimbursement and allow active marketing of the test.

The FDA has already shown it is willing to approve a prostate cancer PET imaging agent on the basis of blinded, independent reads of existing patient scans. In May 2016 it approved Axumin (18F-fluciclovine, Blue Earth Diagnostics) based on two studies that utilised re-reads by three blinded independent readers of scans that had previously been read by on-site readers; in each study the scans were from men with suspected recurrence of prostate cancer based on rising PSA levels. In the first study in 105 men, the Axumin PET scans were compared to histopathology obtained by biopsy of suspicious lesions. In the second study in 96 men, Axumin PET scans were compared to PET scans with 11C-choline.

Telix already has access to data from historical TLX591-CDx PET scans in Europe, and will receive additional scan data from the MSKCC and Endocyte trials. It plans to hold a pre-NDA meeting with the FDA in Q418 where it will propose a Phase III programme for TLX591-CDx that is based on blinded re-reads of existing data. If this strategy is acceptable to the FDA, then management anticipates an NDA submission in 2019. Given that the FDA has already reviewed the manufacturing and QC data in the DMF, there is the potential that TLX591-CDx could be approved by the FDA by the end of 2019 (we model a more conservative timeline with a potential approval in 2020). The estimated cost of a Phase III study and NDA filing based on re-reads is US$2m.

According to the National Cancer Institute, prostate cancer is the second most common cancer in men; it is expected to account for 165,000 new cases and 29,400 deaths in the US in 2018, while 2.9m men in the US are estimated to be living with a diagnosis of prostate cancer. Globocan predicts that in 2020 there will be 1,393,000 cases and 386,000 deaths from prostate cancer worldwide, while in the EU in that year there will be 398,000 new cases and 84,000 deaths. At diagnosis, the disease is localised to the prostate in 78% of cases, while it has spread to regional lymph nodes in 12% of cases, and has metastasised elsewhere in the body in a further 5% of cases (staging is unknown in 4% of cases). While the overall five-year survival rate is very high at 98.2%, the prognosis ranges from a five-year relative survival of 100% for patients diagnosed with localised tumours to only 30% for patients who have metastatic disease at diagnosis.

The PSA blood test is the main tool used for monitoring prostate cancer patients for disease recurrence. An estimated 15-30% of patients experience rising PSA levels (biochemical recurrence) after prostatectomy surgery or radiation therapy. In our view, the main application for TLX591-CDx imaging is likely to be for detecting prostate cancer metastases in men with biochemical recurrence for whom the site of metastasis or recurrence can’t be identified by standard imaging methods such as CT, MRI or bone scintigraphy.

A second potential indication for TLX591-CDx would be to screen newly diagnosed patients assessed as having high-risk disease, based on biopsy results, for metastases to pelvic lymph nodes as well as distant sites. For both of these indications a key aim of the imaging would be to identify the site of metastasis or recurrence so it can be treated by surgery or radiation therapy.

In assessing the addressable market for TLX591-CDx, we bear in mind that while 15-30% of prostate cancer patients experience biochemical recurrence, only a proportion of these cases will be unable to find the metastases by standard methods. On the other hand, some men may have multiple recurrences or may have multiple PSMA scans if the initial scan does not identify any metastases following biochemical recurrence. Weighing up all of these factors, we believe that 50% of the number of men diagnosed each year is a reasonable estimate of the potential demand for PET scans in men with biochemical prostate cancer recurrence.

In regard to the second potential indication in newly diagnosed prostate cancer, we note that 20-30% of men diagnosed with localised prostate cancer present with high-risk tumour characteristics.18 Combining the midpoint of 25% for the incidence of high-risk prostate cancer with our estimate of 50% for biochemical relapse requiring PET imaging, our estimate of the total addressable market is 75% of the number of men diagnosed each year, which is equal to 91,000 scans per year in the US.

A number of targeted agents for PET imaging of prostate cancer have been investigated, in addition to 68Ga PSMA-targeting products. Most interest has focused on 68Ga or 18F-babelled PSMA, 11C or 18F labelled choline and 18F fluciclovine, which is marketed as Axumin.

Exhibit 7 shows that the sensitivity and specificity PSMA and choline PET imaging products for detecting recurrent prostate cancer are equivalent at 85-89%. While the sensitivity of fluciclovine/Axumin is high the specificity is low, which means there will be a high number of false positive reports. For comparison, Exhibit 7 also includes 18F-FDG, which is widely used to detect metabolically active cancers such as lung, colorectal and breast cancers, but is not a sensitive test for prostate cancer, which is less metabolically active than many other cancers.

A key advantage of PSMA imaging agents is that they are more sensitive than other agents for detecting metastases in biochemically recurrent prostate cancer where the PSA levels are low (ie when the tumours are small). In the studies included in a review of prostate cancer PET radiotracers19, the sensitivity of PSMA PET for detecting recurrent prostate cancer metastases in men with low PSA levels (<2.0 ng/ml) was 67%, vs 41% for fluciclovine and 34% for choline. Given it is these small metastases that are the most difficult to detect with standard imaging methods, we expect the increased sensitivity in this patient group will see PSMA PET become the preferred imaging technology for recurrent prostate cancer.

We expect a key driver of uptake of the TLX591-CDx shake and shoot kit will be that it is as well suited to smaller clinics as it is to large hospital sites, because it does not require access to a cyclotron to generate the radioisotope or a centralised manufacturing facility to generate the imaging agent.

While the FDA has not approved any PET imaging agents that target PSMA, it has approved prostate cancer imaging agents based on choline and fluciclovine. The availability of prostate cancer-specific PET imaging is rapidly expanding in the US; as of 2017 it was available at 34 sites – eight using 68Ga or 18F PSMA, eight using 11C or 18F choline and 18 using Axumin20.

Axumin (Blue Earth Diagnostics) was approved by the FDA in May 2016 and by the European Commission in May 2017. It is indicated for PET imaging in men with suspected prostate cancer recurrence based on elevated PSA levels following prior treatment, and has subsequently been included in the National Comprehensive Cancer Network clinical guidelines for this indication. Axumin contains the 18F-labeleld synthetic amino acid analogue fluciclovine. 18F fluciclovine is taken up by amino acid transporters such as ASCT2 and LAT-1 and accumulates to a greater extent in prostate cancer cells compared with the surrounding normal tissues. Blue Earth reported global sales for the year to September 2017 of £15.8m (~US$21m), mostly from the US market (vs £0.5m in the preceding year). Sales in the six months ending March 2018 were £23.5m (US$30m). Blue Earth sold 3,700 Axumin units in the US in Q417, 5,000 units in Q118 and 6,000 units in Q218.

The FDA approved manufacture and use of a second agent, 11C choline, for PET imaging in recurrent prostate cancer at the Mayo Clinic in 2012. Choline has subsequently been approved at three other hospitals as well as for manufacture and supply by the radio-pharmaceutical manufacturer Zevacor Molecular. Choline is an essential component of cell membranes and accumulates in tissues with high cellular proliferation.

Despite the fact that the FDA has not approved any PET imaging agents that target PSMA, we noted above that as of 2017 at least eight institutions were providing a PSMA-PET imaging service as an investigational agent. A number of other companies are also developing PSMA-based PET imaging products, so Telix may face direct competition from an approved PSMA imaging test in the US, in the future. Four potential competitors are described briefly below.

Progenics (NASDAQ: PGNQ) has completed enrolment in a Phase II/III study of a PSMA-based PET imaging agent 18F -DCFPyl, which was developed at Johns Hopkins. The study enrolled 383 patients with localised high-risk prostate cancer or recurrent or metastatic disease and is expected to report top-line data in Q418. A second Phase III in patients with biochemical recurrence is expected to commence in Q418.

The UK-headquartered, privately owned company Theragnostics has developed a room-temperature ‘shake and shoot’ kit for 68Ga -THP-PSMA (also known as THG-001), which sounds similar to the ANMI technology used with TLX591-CDx. Theragnostics recruited the first patient in a Phase II study in high-risk or recurrent prostate cancer in June 2018, but does not appear to be supplying the product on a commercial basis in the US.

Novartis also recently entered the field when its recently acquired subsidiary Advanced Accelerator Applications signed a licence agreement with US-based private company Cancer Targeted Technology to develop and market CTT1057. CTT1057 is an 18F -based PSMA tracer, which recently completed a Phase I clinical trial.

In May 2018 Blue Earth diagnostics in-licensed global rights to a family of 18F-PSMA MTR imaging agent from Scintomics. The acquisition of this preclinical PSMA imaging programme suggests to us that Blue Earth expects PSMA diagnostics to play a key role in prostate cancer imaging.

For completeness, we mention much older PSMA imaging technology ProstaScint, an 111In-labelled mAb that targets an intracellular PSMA epitope, which the FDA approved in 1996. ProstaScint (Aytu Bioscience; NASDAQ: AYTU) uses an older imaging technology, SPECT scan, which is much less sensitive than the PET imaging technologies described above (Exhibit 7). ProstaScint is approved as a diagnostic imaging agent for detecting lymph node metastases before surgery in patients with high-risk, biopsy-proven prostate cancer, and for post-surgery patients with rising PSA and negative or equivocal standard metastasis evaluation. However, sales have been modest and Aytu intends to withdraw ProstaScint from the market in the US at the end of 2018.

US Medicare reimburses Axumin at US$3,900 per dose, while 11C choline is reimbursed at US$5,700 per dose21. Medicare typically reimburses drugs on a pass-through basis for three years, after which the cost of the drug is packaged into the overall procedure cost. We model average revenue to the JV of US$3,500 per procedure.

Telix is developing TLX591, a 177Lu-labelled MTR therapeutic for treating prostate cancer. TLX591 is based on an enhanced version of the huJ591 mAb.

Telix has a research collaboration and option agreement with Atlab Pharma, a privately owned French biotech, which holds the right to the huJ591 antibody for use with 177Lu and certain other combination therapy rights. Atlab in-licensed the rights from Cornell. Telix is exercising its option to acquire Atlab for ~US$10m.

Atlab conducted a number of clinical trials of a single cycle of treatment with 177Lu-J591 (which Atlab called ATL101), in conjunction with Weill Cornell. When 177Lu-J591 was administered as a single infusion, the MTD was 70 mCi/m2 and 11-12% of subjects treated at 65 or 70 mCi/m2 achieved at least a 50% reduction in PSA count.

The tolerability was improved in a subsequent study by splitting the total dose into two fractions administered two weeks apart. The total dose tolerated as two fractions was higher at 80-90 mCi/m2 (2x40 mCi or 2x45 mCi with the option of GCSF white blood cell growth factor). The higher total fractionated dose was also more efficacious; 21% of subjects treated at the RP2Ds experienced a 50% reduction in PSA, including 29% of subjects treated at the highest dose of 2x45 mCi22.

In a separate study when a single cycle of fractionated 177Lu-J591 was administered in combination with a multiple cycles of docetaxel chemotherapy, 11/15 (73%) of subjects achieved at least a 50% reduction in PSA levels23.

In each study of 177Lu-J591 the dose-limiting toxicity was reversible myelosuppression, a condition where bone marrow activity is decreased, resulting in fewer red blood cells, white blood cells, and platelets.

In the Phase I study of dose fractionated 177Lu-J591, 36/49 (74%) subjects treated in all six dose cohorts had grade 3/4 haematological toxicity, while 19/33 (58%) of subjects treated at the RP2D had the more severe grade 4 haematological toxicity. In the docetaxel combo study, 20% experienced grade 4 neutropenia without fever, while 13% experienced grade 4 thrombocytopenia.

Overall, the studies showed that fractionated 177Lu-J591 is tolerated with subsequent PSA declines and reversible myelosuppression. The falls in white blood cell counts were severe but did not result in fever (so-called febrile neutropenia, which can be life-threatening). While the haematological toxicity/myelosuppression increases the risk of bleeding events or infection, it does not otherwise impair patient quality of life, unlike, for example, nausea, diarrhoea or neuropathy often associated with chemotherapy. However, neutropenia can result in a delay in subsequent chemotherapy treatment if the patient progresses despite MTR therapy.

While TLX591 is based on ATL-101/177Lu-J591, Telix has made a number of modifications to optimise its performance. It has been re-humanised, stability/affinity optimised, and re-engineered to have better properties for radioactive drug use.

In general, at high doses radio-labelled mAbs have higher haematologic toxicity than small molecules because of their long circulation time (around one week serum half-life) compared to around a 48 hour clearance from radio-labelled peptides. To address this, Telix has modified the constant region of the TLX591 mAb to stop it binding to the Fc receptor on white blood cells. As a result, TLX591 is cleared from the bloodstream much more rapidly, which is expected to result in less haematologic toxicity.

While the company anticipates that TLX591 will have better efficacy and fewer toxic side effects than 177Lu-J591, this will need to be tested in clinical studies.

Drug material for toxicology and human biodistribution studies is expected to be ready by the end of 2018. The company is planning to conduct initial biodistribution/pharmacology studies for TLX591 in Australia. It is on track to initiate an Australia/US multi-centre Phase II study in mid-2019.

Although TLX591 (177Lu-J591) has shown encouraging efficacy in clinical studies, Telix may face competition in this space from Endocyte’s small molecule therapeutic 177Lu-PSMA-617, which also binds to the PSMA receptor and reported high response rates and low toxicity in a Phase II study in metastatic castration resistant PC (mCRPC). Preliminary data reported by Endocyte at ASCO in June from the 50-patient Phase II study showed that 62% of subjects experienced a PSA decline of at least 50%. The treatment was well tolerated, with the most common side effect being grade 1–2 dry mouth reported by 68% of subjects (66% grade 1, 2% grade 2). The occurrence of grade 3–4 hematologic toxicity was low; there were no cases of grade 4 neutropenia and only 6% of subjects experienced grade 3 neutropenia. Patients were dosed every six weeks for up to four cycles.

In June 2018 Endocyte enrolled the first patient in a 750-patient global Phase III study of 177Lu-PSMA-617 in PSMA-positive mCRPC. The dose in the Phase III study is 7.4GBq every six weeks for up to six cycles. Endocyte is using Telix’s TLX591-CDx kit to screen patients for enrolment into the trial.

In 2013 the FDA approved Xofigo (Bayer), a bone-targeted alpha particle called radium 223, which leads to longer overall survival in men who have with symptomatic mCRPC with bone metastases but no known metastases in visceral organs. In a Phase III study, Xofigo improved survival by 3.6 months (14.9 months versus 11.3 months; HR, 0.70).24

At its launch in 2013, Xofigo was priced in the US at US$69,000 for a course of six injections four weeks apart. We estimate that, since then, the manufacturer’s list price has increased to currently be ~US$91,000 for a course of treatment with Xofigo. We base this estimate on the wholesale acquisition cost or WAC of US$131.85 per mCi, the recommended dose of 1.35 mCi/kg and average weight of a US male aged 60-80 of ~85kg. Worldwide sales of Xofigo in 2017 totalled €408m, including €242m in the US (US$492m and US$292m, respectively). Bayer expects peak sales of Xofigo to be in excess of €1bn (US$1.2bn), although the increased death rated reported in patients receiving Xofigo plus Zytiga in a recent Phase III study in chemotherapy-naïve patients with no symptoms or only mild symptoms may limit future sales growth. In the US there are also older FDA-approved treatments utilising radioactive isotopes to reduce pain from prostate cancer that has spread to the bones, called samarium-153 (Quadramet) and strontium-89 (Metastron).

While these bone-targeted radioisotopes are useful because prostate cancer commonly spreads to bone, those drugs cannot treat tumours in other sites such as in the prostate, lymph nodes, or lung. Given that TLX591 could be used to treat patients with soft tissue metastases as well as bony metastases, it would be applicable to a greater proportion of prostate cancer patients than Xofigo.

In October 2017 Telix acquired the privately held German company Therapeia, which held and developed the background technology to TLX101.

TLX101 is a radiolabelled amino acid analogue (4-[131I]iodo-L-phenylalanine, also known as 131I-IPA), which is a small molecule that rapidly crosses the blood-brain barrier. TLX101 is taken up by the membrane-bound L-amino acid transporter 1 (LAT1), which is over-expressed in gliomas (brain cancers) including glioblastoma (GBM). This allows the tissue-specific accumulation of the 131I therapeutic radionuclide inside the cancer cells. The 131I radiolabel is used because it readily crosses the blood brain barrier, whereas a charged radiometal such as 177Lu does not.

In addition, preclinical studies show that unlabelled IPA is a radio-sensitiser, ie, it increases the sensitivity of cells to radiation. In studies in an animal model where GBM tumours were growing in the brain of rats, combining TLX101 with external-beam radiation therapy (EBT) was much more effective than either therapy alone25.

Telix also acquired TLX101-CDx which uses IPA labelled with the radionuclide 124I for PET imaging, which Telix will use as a research tool to study the pharmacology of TLX101. PET imaging of patients with commercially available MTR diagnostics that cross the blood brain barrier and are taken up by LAT1, such as 18F-FET or 18F-flucyclovine (FACBC), could potentially be used to screen patients for enrolment in the upcoming clinical trials of TLX101

IPA labelled with a different radioisotope, 123I, has previously been investigated for SPECT imaging of high grade gliomas. It showed uptake and prolonged retention and promising diagnostic performance. In an academic study in 100 patients with glioma the sensitivity to detect glioma was 88% with specificity for glioma of 95%. The sensitivity for high grade gliomas (mainly GBM) was 93%, therefore the vast majority of GBM patients should be eligible for therapy with TLX101.

A total of 11 patients have been treated with 131I-IPA (TLX101) on a compassionate use basis in two institutions in Germany. Baum et al. (2011)26 reported that among two patients with low-grade glioma treated with TLX101, one experienced a reduction in tumour volume over 10 months, with progression occurring 10 months after therapy. The second patient showed stable disease on MRI and PET during the three-month follow-up period. There was no clinically detected toxicity. Company data shows that, subsequent to the published study, the second patient was treated with 131I-IPA + 68 Gy EBT, which resulted in a measurable (40%) reduction in tumour volume and clinical improvement including the cessation of epileptic fits. In addition, a patient with GBM who was treated with 6 GBq 131I-IPA + 68 Gy EBT achieved histologically confirmed tumour eradication,27 although the patient died eight months later from (unrelated) gastrointestinal bleeding.

The company intends to conduct the IPAX-1 Phase I/II trial of TLX101 in at least 35 (potentially up to 55) relapsed GBM patients in connection with external-beam radiation therapy. The study will begin with a dose escalation component in ~20 patients. The efficacy assessment will be based on post-treatment imaging and is expected to read out by the end of 2019. If the trial is successful it would provide sufficient clinical data to decide whether to proceed to a Phase III trial. TLX101 has orphan drug designation in the US and Europe.

The twin aims of combining TLX101 MTR with EBT are to exploit the radio-sensitising properties of TLX101 and to maximise the dose of radiation delivered to the tumour without increasing the incidence of dose limiting toxicities. The dose of radiation delivered to the tumour can be maximised because the side effects of MTR and EBT affect different tissues. The side effects of EBT are due to the effect on other parts of the brain in the path of the X-ray beam, while the side effects of MTR are due to its accumulation in non-target tissues. Because there is very little overlap in the tissue distribution, combining the two modalities of radiation therapy can increase the total dose of radiation that can be delivered safely to the tumour.

GBM is a compact market with most patients managed by specialised centres and clinical trials in GBM are typically modest in size. For these reasons, the company’s current strategy is to retain full ownership of TLX101 and develop and commercialise it itself in the US, if approved.

According to the National Cancer Institute, brain and central nervous system (CNC) cancer is expected to account for 23,900 new cases and 16,800 deaths in the US in 2018. Globocan predicts that in 2020 there will be 300,400 cases and 225,900 deaths from brain and CNS cancer worldwide, while in the EU in that year there is predicted to be 46,700 new cases and 36,300 deaths. The prognosis is poor regardless of the stage at diagnosis, with an overall five-year survival rate of only 33.2%.

GBM is the most common and most aggressive primary malignant tumour of the brain and spinal cord. Approximately 11,000 patients are diagnosed with GBM each year in the US, representing 46% of all brain and CNS cancers. GBM tumours are characterised by invasive and diffuse growth, which makes complete surgical removal difficult. Standard treatment for GBM entails maximal surgical resection of the tumour followed by radiotherapy with concurrent chemotherapy with temozolomide followed by adjuvant chemotherapy with the same drug to treat the residual infiltrative component of the tumour. Despite this aggressive treatment, the disease invariably returns resulting in a five-year survival rate of only 5%.28 Options for treating recurrent GBM include bevacizumab (Avastin), alternating electric field therapy (Optune), surgery and EBT29. Improved treatments are urgently needed.

Telix is subject to typical biotech company development risks, including the unpredictable outcome of trials, regulatory decisions, success of competitors, financing and commercial risks. Our model assumes that TLX250 and TLX591 will be out-licensed; therefore, our valuation is sensitive to potential licensing timing and actual deal terms. While Telix is commencing commercialisation of TLX591-CDx, it is still mainly a mid-stage drug developer, therefore in the foreseeable future most value creation will depend on successful R&D progress and any potential partnering activities, although the timing of licensing deals is typically difficult to forecast. If the FDA requires Telix to conduct a prospective clinical study of TLX591-CDx rather than a blinded review of existing scans, then that could delay filing for full marketing approval by one to two years. Predicting utilisation of PSMA-PET imaging such as TLX591-CDx in prostate cancer is challenging; we model a scenario where it is used as a second-line test when metastases cannot by detected by CT or MRI scans, but if it becomes a first line test in prostate cancer patients with biochemical relapse, or if patients typically undergo repeated PET scans, the market could be significantly larger.

Our initial valuation of Telix is A$303m based on a risk-adjusted discounted cash flow model, which includes our estimates of the future milestone payments and royalty streams for TLX250 and TLX591, plus profits from commercialisation of TLX250-CDx, TLX591-CDx and TLX101, as listed in Exhibit 8. We have extended our cash flow forecasts out to 2037 (supported by 12 years of biologicals market exclusivity in the US and 10 years in Europe) but have not included any terminal valuation. We assume a long-term exchange rate of US$0.76/A$ and apply a 12.5% discount rate. We assume that the Atlab acquisition completes as expected in Q318 and that the US$10m consideration is paid through that issue of 20.5m Telix shares. Our valuation is equal to A$1.39 per share on an undiluted basis but including the Atlab acquisition shares and A$1.36/share after diluting for the 10.6m options on issue (exercise price 85c; all of the options would be in the money if the stock was trading in line with our valuation).

We estimate that the current list price of the prostate cancer therapeutic radiopharmaceutical Xofigo in the US is US$91,000 for a course of six injections, as we explain on page 16. Allowing for discounts and rebates, we assume that the net sales price of Telix’s MTR therapeutics is US$70,000 per patient in the US, with a 30% discount applying in other markets.

Exhibit 8 shows our market assumptions for TLX250, TLX591 and TLX101 imaging and therapeutic product (TLX101 therapeutic only) and the rNPV for each product. We have offset the risk-adjusted trial cost against revenue for each indication.

Deal terms are based on relevant benchmarks over the last few years (sourced from EvaluatePharma and the industry group BIO). There are few directly comparable deals, ie mid-stage, MTR therapeutics and imaging agents for cancer. Therefore, we looked firstly at deal terms within the targeted therapy space, which included average upfront/milestones of US$127m/US$788m (Exhibit 9). Additionally, based on data in a report produced by BIO, we calculated that among 124 global licencing deals for Phase II therapeutics from 2013 to 2017 (for all diseases) the average upfront/milestones were US$46m/US$281m. Averaging these two data sources, we assume upfront/milestones of US$86m/US$535m for a licence deal for Telix’s product pipeline. We assume half of those milestone payments (US$268m) are for clinical and regulatory milestones and half are sales-based milestones. We split the US$86m upfront and US$268m clinical and regulatory milestones between the TLX250 kidney cancer therapeutic and the TLX591 prostate cancer therapeutic, adjusted with a 20-70% probability and weighted according to peak sales.

We do not include the potential sales-based milestones in our forecasts, and instead model a 15% royalty rate for the Phase II products (Exhibit 8, above, shows our assumed net royalty due to Telix after deducting trailing royalties).

Telix was incorporated on 3 January 2017 and reported an operating loss for the year ending 31 December 2017 of A$6.4m. Expenses associated with R&D projects were A$3.0m, personnel expenses were A$1.3m and administration and consulting costs totalled A$2.3m. Our total operating loss estimates for 2018, 2019 and 2020 grow to A$13.0m, $17.9m and $15.8m, respectively, mainly due to increased R&D expenditure (detailed in Exhibit 8) as Telix ramps up its clinical trial programme, partly offset by the Australian government’s R&D rebate scheme. Telix has received an advance/overseas R&D tax finding totalling A$55.2m regarding the eligibility for the R&D rebate of overseas R&D expenditure that is essential to Telix’s programmes but cannot be executed in Australia by Australian vendors and service providers.

Telix had A$42.0m cash and equivalents at 30 June 2018 as compared to A$48.7m at the beginning of 2018. Although Telix has substantial cash reserves and is funded into 2020, we estimate that it may need additional funding of A$4m in 2020, which we include as indicative long-term debt. The actual funding requirement (if any) will depend on the rate of R&D expenditure and the timing of licensing transaction. Our estimated funding requirement is based on our assumption that Telix out-licences, TLX-250 and TLX591 in a single transaction in 2021 by which time the results of a Phase II study of TLX591 in prostate cancer are expected to be known. However, if it enters a licensing transaction in 2020 it may not need any additional funding.