Over the weekend, a reader (a scientist in translational medicine) kindly sent me the link to a paper on PARP inhibition and asked:

"Is this a sign of the new wave of oncology drug development? Rather than basing treatment on cancer tissue type (eg. breast, prostate, colorectal), the underlying genetic mutation regardless of tissue of origin will be used for targeted agents. Imagine a randomized Ph2 trial of all BRCA1/2 deficient patients vs. non-mutated patients. Of course, this same rationale would have not worked for K-ras mutation patients with EGFR inhibitors in NSLCLC vs. Colorectal."

My short answer is basically, yes.

It may, however, take a while for this to evolve fully clinically, but you can already see the movement starting at AACR meetings, where they organise topics around pathways.

Unfortunately, at ASCO and ASH things are still discussed in cancer specific sessions, which I increasingly find more difficult to follow and plan for.  For example, suppose I'm interested in pipeline PI3-kinase inhibitors.  There might be something at 8.45am in North Hall A on one tumour type, but another at 9.00am on the other side of the convention centre in Room 330A in the East Building on a different cancer.  Ugh, no wonder we all joke about the #blisterwalks and increasingly lack of time for networking at conferences – we're all too busy dashing between sessions in highly frazzled, rather than leisurely fashion!

I do believe the translational medicine and scientists are right in the pathway over tumour approach though and have been advocating for a pathway driven approach for some time on this blog.  If you haven't read it yet, the short four part series on pathways and treatment in lung cancer earlier this year may be useful (see posts here, here, here and here) for examples of how lung cancer specialists (translational scientists and clinicians together) are leading the way in pioneering this concept.

In the example my correspondent gave, PARP inhibitors are being tested in BRCA1 and 2 positive tumours for olaparib (AstraZeneca), i.e. breast and ovarian cancers, whereas looking at the clinical trials database, Sanofi-Aventis are taking a more traditional approach with iniparib (BSI-201) and looking at different tumour types without the BRCA mutation.

Meanwhile, Abbott are also taking a very creative approach with their PARPi, veliparib (ABT-888), by not only looking at BRCA-positive cancers but also getting involved in the I-SPY2 neoadjuvant trial in breast cancer, which was a very smart move in my opinion.  You can read more about the agents in that groundbreaking study protocol from the interview with Dr Sue Desmond-Hellmann.  It's a great example of seeking to treat women diagnosed with early breast cancer with different therapies based on their underlying biology of their tumour, thereby enabling us to see what works and what doesn't earlier than we would normally.

It's going to be interesting to see how all these different R&D strategies work out commercially.  For the record, I think olaparib may well be the first targeted agent to show benefit in patients with cancers that result from BRCA1 or BRCA2 gene mutations.  That's quite exciting in itself, and hopefully, AZ have also learned considerably from the mistakes made with gefitinib (Iressa).  Whether the specific targeted and scientific approach will win over a more commercial strategy, we'll have to wait and see.

The advantage of a specific targeted strategy is that if you get the biomarker right, the people who are most likely to respond to a therapy will more likely receive it plus the response rates and outcomes, in theory, should be better.  Those who are less likely to respond are spared of the systemic exposure and false hope.

Some marketers will also argue that it may also, however, limit your market commercially, but my argument is that a highly targeted agent will actually get more prescriptions in the long run because it will be easier for an oncologist to choose therapy accordingly and good results reinforce usage and loyalty. Standing out from the crowd with superior efficacy in a well defined population is a much more elegant approach to this marketer

In a catch-all strategy, you may believe that by targeting a bigger market you will win more commercially, but the risk is that the response rates will be lower in a heterogeneous population (the wins and losses will cancel out to some extent). Thus if a competitors more targeted approach actually works, they win and you have no biomarker to answer with to allow oncologists to choose between the agents. Increasingly, they are reluctant to treat everyone willy nilly because of high costs, whether in terms of patient co-pays for oral therapies or more draconian restrictions for iv therapies from insurers.

Developing targeted agents is very different from traditional chemotherapy. Oncologists now want to know who is most likely to respond to a given therapy and treat those subsets accordingly rather than randomly treat and expose everyone to not insignificant systemic side effects of therapy.  One of the best concepts we have right now is to have multi-functional research and clinical practice that marries identifying the key pathways and mutations with targeted therapies designed to knock out or shutdown the aberrant cell signalling.

We have a ways to go, but the tide is already turning.

Tutt A, Robson M, Garber JE, Domchek SM, Audeh MW, Weitzel JN, Friedlander M, Arun B, Loman N, Schmutzler RK, Wardley A, Mitchell G, Earl H, Wickens M, & Carmichael J (2010). Oral poly(ADP-ribose) polymerase inhibitor olaparib in patients with BRCA1 or BRCA2 mutations and advanced breast cancer: a proof-of-concept trial. Lancet, 376 (9737), 235-44 PMID: 20609467