|
|
|
|
||
Re: Here is Yervoy + Opdivo combo for NSCLC sBLA data Which I got from CTMX boardCytomX' Translational Data On CX-072 Point Towards A Positive Readout From CX-2009
About: CytomX Therapeutics (CTMX), Includes: ABBV, AMGN, BMY SummaryInitial data generated by CX-072, CTMX' PD-L1 prodrug, have failed to woo investors due to the difficulty involved in interpreting data from small dose escalation cohorts across cancer types. However, translational analysis presented at SITC provides meaningful confirmation that CX-072 behaves as intended, thereby validating the underlying 'probody' technology for use in humans. While I expect the growing CX-072 dataset to clarify the drug's intrinsic value, I believe that translational implications for the company's CD166-targeted conjugate are far more impactful. The company is currently dosing CX-2009 up to 10mg/kg, which together with the CX-072 translational data suggests absence of harmful binding to CD166 expressed in healthy tissue. This appears to de-risk CX-2009 ahead of an interim readout in 1H 2019 which should cement this unique conjugate's potential for the treatment of all CD166-expressing solid tumors. This idea was discussed in more depth with members of my private investing community, Second-Level Investing . Get started today » IntroductionSince first covering CytomX (CTMX) back in January, the company's stock has been on a wild ride while the company has made significant strides towards the validation of its game-changing 'probody' platform: Annotated 1-year chart created by the author. In my initial public analysis of the company's prospects, I summed up their approach as follows:
Another way to think about 'probodies' is that they are 'antibody prodrugs', designed so they are only activated once they reach the tumor microenvironment, while remaining inactive in circulation. In January, I analyzed the extensive and very convincing preclinical work conducted by the company but I had no meaningful clinical data at my disposal to derive conclusions as to the probability of success ('POS') of the company's ambitious pipeline. In the meanwhile, the company has presented preliminary data from its lead molecule, the PD-L1 probody CX-072, across mono-therapy and combination therapy cohorts, and at different dose levels. Most recently, the company has provided translational analysis at the medical conference SITC which provides investors with a highly relevant validation of the probody platform. Before I delve into the data, a quick recap as to meaningful progress made since my article in January:
On the financial side of things, the company had $465m in cash, cash equivalents and short-term investments as of September 30, up from roughly $350m earlier this year. Currently, R&D and SG&A expenses clock in around $100m on an annual basis, meaning the company should be funded through 2022/2023 depending on how expenses ramp up & to which extent additional milestone payments from existing partners Abbvie (ABBV), Amgen (AMGN) and Bristol-Myers Squibb (BMY) are realized. A discussion of the CX-072 clinical data to dateCytomX are evaluating their PD-L1 probody CX-072 in a clinical program that involves dose escalation in mono- and combo-therapy as well as expansion cohorts. To date, the data generated by the dose escalation portion have served to identify a recommended dose - 10mg/kg - for further evaluation in the expansion cohorts while providing initial indications of safety & efficacy. Observers, myself included, were not particularly impressed with the initial CX-072 readouts but this needs to be offset by the difficulty associated with interpreting such a small dataset across doses & tumor types. CX-072 mono-therapy An analysis of the most recent dataset derived from the mono-therapy, dose escalation portion of the program reveals pharmacokinetic ('PK') properties that are relevant to the translational considerations we will discuss further down:
Here, CytomX have compared:
What this shows is that:
With regards to the study population so far, we are looking at a total of 46 patients across a wide variety of solid tumors who have received a median of 3 prior treatments (range 1-13). Only 13/46, or 28% of patients treated so far, had tumors exhibiting high levels of PD-L1: The heterogeneity of the tumors being treated, their low degree of PD-L1 expression and the variety of CX-072 doses being evaluated here make it futile to derive definite conclusions with regards to CX-072's efficacy and safety profile - such conclusions should be reserved until a meaningful amount of data from the mono-therapy & combination therapy expansion cohort at 10mg/kg is available. A few things can be noted on the safety front nevertheless:
Moving onto preliminary efficacy, 10mg/kg emerges as an ideal dose (additional discussion in the next section): CX-072 combo therapy with Yervoy Initial CX-072 safety & efficacy data in combination with Yervoy (BMY), an immunotherapy ('IO') drug targeting CTLA-4, is potentially even more important than the mono-therapy data, seeing how the strategic opportunity for CX-072 is to replace other PD-1/PD-L1 antibodies as the backbone of choice for IO combination therapy going forward. Avid followers of the oncology space will know that Bristol-Myers Squibb have submitted a supplemental biologics license application ('sBLA') to the FDA in order to establish their Yervoy + Opdivo combination as a frontline treatment of advanced non-small cell lung cancer ('NSCLC'). This sBLA was filed on the basis of data clearly favoring this IO combo over chemotherapy:
CX-072 + Yervoy can be thought of as a next-generation version of the Opdivo + Yervoy combo for which Bristol are seeking frontline approval in NSCLC, and CTMX investors should keep in mind that Bristol is one of CTMX' long-standing partners. The most recent data presented from the dose escalation portion of the Yervoy & CX-072 combo arm compiles data from 20 patients with a median 3 prior treatments, in-line with the CX-072 mono-therapy arm:
As with the monotherapy dose escalation portion, the combo was tested in a wide variety of cancers. Median duration of treatment in the relevant 10+3mg cohort was only 1.3 months, reflecting the progressive nature of dose escalation: Keeping this limitation in mind, so far, the safety profile of 10+3 is looking very favorable, with only 1/7 patients experiencing a grade 3/4 TRAE: No immune-related adverse events have been noted in the 10+3 cohort as of yet, a stat that needs to be considered against the short duration of treatment to date, but which is encouraging nevertheless given the combined effect of two checkpoint inhibitors: For reference, 15% severe to fatal immune-mediated adverse reactions are noted in Yervoy's label, when used as mono-therapy at 3mg/kg: A commentator on twitter has argued that the adverse event profile as noted on Yervoy's label stems from trials in melanoma, a highly immunogenic tumor, whereas the ongoing Yervoy & CX-072 combo trial is conducted in difficult-to-treat, 'cold' tumors - and that data on Yervoy's label is not directly comparable to the PROCLAIM trial. Point well taken - we won't be able to make any formal conclusions until we have a more substantial dataset from the expansion cohorts. When we do, we'll be able to evaluate those data against results of Opdivo + Yervoy in NSCLC noted earlier. CTMX feel encouraged by the safety dataset of CX-072 + Yervoy to date and have initiated enrollment in an aggressive 10mg+6mg cohort: Understanding the CX-072 translational dataCytomX hosted a webcast over the weekend to discuss translational findings from their ongoing PROCLAIM trial of CX-072 in solid tumors, presented at the SITC conference. These findings provide meaningful validation of the underlying 'probody' technology and they yield additional insights, such as 10mg/kg likely being an ideal dose. Even more importantly in my opinion, these translational data from the PROCLAIM program can be correlated with the fact that the company is successfully proceeding to dose-escalate their CD166-targeted probody-drug-conjugate CX-2009 at 10mg/kg. Together, these data points & observations further strengthen my conviction that the company will post highly significant safety & efficacy data from the CX-2009 program in H1 2019. I have previously elaborated on the truly extraordinary commercial prospects of a 'universal' anti-cancer biologic like CX-2009. In a nutshell, the probody platform enables CTMX to target antigens like CD166 or the transferrin receptor-1 which are widely expressed across both healthy and cancerous tissues, thus making them off-limits for traditional antibodies & antibody-drug-conjugates ('ADCs'). CD166 is expressed far more ubiquitously and evenly than commonly targeted cancer antigens such as HER2. Source Let me walk you through the relevant scientific poster, which can be found on the CytomX website. The company used tumor biopsies pre- and post-treatment and applied a variety of advanced investigational methods - which are explained in greater detail in these slides - to determine:
The above chart suggests that 10mg/kg yields significant and consistent levels of activated CX-072 expression inside solid tumors. 30mg/kg, while efficacious, might be too high a dose in terms of tolerability and does not necessarily yield much better results in real life. This is further cemented by measurements of PD-L1 target occupancy by CX-072, in other words a measurement of how much of the tumor-expressed PD-L1 is bound by activated CX-072 probody, which show that target occupancy is strong at 10mg/kg, whereas only 3 out of 8 biopsies from patients treated at 3mg/kg showed detectable, activated CX-072. The company conducted theoretical modeling of target occupancy rates at different doses, which overlapped very nicely with real-world data at the 10mg & 30mg doses. The company conducted testing in a mouse model which yielded findings that can be translated & compared to results in humans at 10-30mg/kg. Not all, but the majority of tumors, showed an increase of immune cell penetration following treatment with CX-072. While the sample size presented here is small, 10mg/kg again appears to be the most promising dose. Finally, a measurement of relevant mRNA biomarkers of T-cell activation in a patient subset shows highly significant expression in a 10mg/kg patient. Taken as a whole, these findings - especially the fact that CX-072 remains masked in circulation, as discussed at the beginning of this article - strongly suggest that CX-072 is behaving per its design, from which we can deduce that the probody platform as a whole is viable, given that there is little to no variability with regard to the peptide 'masks' used between different probodies. What to expect from the CX-2009 readout in 1H 2019The CX-2009 program is enrolling across at least 7 cancer types and dosing has progressed to 10mg/kg. Source Since:
I deduce that:
As a result, I expect the initial CX-2009 readout in 1H 2019 to show highly encouraging safety and efficacy signals. Per my own calculations, CX-2009 has the potential to address a $16bn/year opportunity across the seven cancer types the company is tackling in the dose escalation arm. That is because CD166 is so widely expressed and there remains a serious unmet need in many of the cancer types listed. The major limitation to ADCs has traditionally been an unfavorable safety profile, but recent advances in protein engineering have improved the PK/PD of these drugs, and the probody platform provides the added benefit that previously off-limit antigens such as CD166 can be targeted. If / when CTMX report favorable data from CD166, and subsequently from their partnered program targeting the transferrin-1 receptor, they will have unlocked an incredible value proposition. CX-2009 has disruptive potential in oncology, and I believe CTMX will be acquired down the line - possibly by one of its current Big Pharma partners - as these datasets expand. |
return to message board, top of board |