Over the past several weeks, coincident with the opening of a new large clinical trial and some publicity associated with that, several people have asked me here about a lung cancer vaccine called Lucanix (full name belagenpumatucel-L, so a near guarantee that nobody will call this anything but Lucanix). My initial response was that I knew essentially nothing about it and more or less implied that there must not be much to it if there was really no buzz about it within the lung cancer community. But that’s a very simplistic view that goes against one of the issues I feel strongly about, namely that hype isn’t the same as promise. There are enough approaches that I’ve been critical of despite “buzz” (RFA, proton beam radiation, celebrex, just to name a few I’ve minimized over the past few weeks) that it would be completely hypocritical for me to suggest that a lack of attention from the media or lung cancer establishment means that a treatment isn’t worth pursuing. I’ve seen over the years that the interest in a treatment is a combination of the favorable findings from early work, the public relations machine and marketing muscle of the sponsor company, how explainable it is (we like a good “story”, so it’s easier to explain that something blocks EGFR or is antiangiogenic than that something is an aurora kinase inhibitor), and certainly some luck about what just happens to catch on. I’ve also been following some treatments that seem to get little traction despite quite encouraging results (talactoferrin alpha, for instance), probably because they’re products of small companies, and/or have been studied by people who don’t immediately attract media attention.
So it was only fair for me to review the available information about Lucanix and provide a more informed opinion. T0 their credit, folks at NovaRx, the company working on Lucanix, saw my comments announcing my intention to write a more educated piece, and they proactively sent me several files with background information about the company, Lucanix, and the trial being conducted now. After several attempts to connect, I spoke earlier today with Ms. Carissa Schumacher, Director of Corporate Development, and Dr. Habib Fakhrai, President; both were eager to discuss the promise and challenges of the early work with Lucanix and the new trial moving forward. I’ve never had the leadership of a company provide more enthusiastic and open communication, so kudos to them for that.
First, what is this vaccine? There are two main types of cancer vaccines. One type is called autologous and is generated from the individual patient’s tumor, which has the advantage of being tailored to the individual, but it requires tumor to be collected from a patient by a separate procedure. Lung cancer tumor tissue isn’t always accessible, so it’s a problem and involves some degree of risk to obtain tissue to generate the vaccine, which is also a time-consuming process. The other type is called allogeneic, which means that it’s made by other tumor cells, not those from the patient. Lucanix is a vaccine derived from four different lung cancer cell lines (two adeno, one squamous, one large cell), so the immune response would presumably be against proteins on any of a number of different lung cancers. This is combined with an “antisense” gene modification against transforming growth factor-B (or TGF-beta), which is a very complicated idea, but the net result is that this helps reinstate immunity against cancer cells after they had disarmed the body’s surveillance system in the immediate area around the tumor with this protein.
This work was first reported at an oral presentation at ASCO in 2006 (abstract here) and then published in the Journal of Clinical Oncology, one of our leading journals, that same year (abstract here). It included 75 patients with anywhere from stage II to IV NSCLC, although nearly all had stage IIIB or IV disease, and most had received prior therapy. Patients were then randomized into three different dose levels of the vaccine, which was administered into the skin (intradermal) once every 1-2 months. Not surprisingly, the vaccine was associated with minimal side effects. While fatigue, pain, and pulmonary symptoms were reported, there was no real indication that this was from the treatment and not the cancer itself. Two patients had significant issues felt possibly or probably related to the vaccine: one had fairly significant pain and swelling around the injection site on one occasion, and another developed chronic myelogenous leukemia (CML) 5 months after 16 full cycles of the vaccine — it remains unclear if the vaccine caused this or whether this was just a coincidence.
In terms of responses among patients with measurable disease, 6 were seen among 40 patients, with 5 in the middle or highest dose. A total of 59% had no progression after 16 weeks on treatment. However, looking at several of their highlight responses, I’ve seen far more convincing responses, and I just don’t feel completely confident that these aren’t improving pneumonitis, infection, or some other non-cancer process, at least in some of these patients. Apparently the physicians involved in the trial were convinced that these were true responses, but I’d really like to see a broader array of data to really be convinced that this can lead to tumors shrinking.
Survival certainly appeared promising, particularly compared to historical data of what you’d expect for previously treated patients. Median survival was 14.4 months on the trial, and it was actually 15-16 months in the two higher dose cohorts, in which the two year survival was about 40%:
I do find it hard to explain why the lower dose would do visibly better than the middle dose, and my honest sense is that this is because the trial involves a small number of patients, so random chance of the particular characteristics of a few patients and their cancers play a large role. I’ve been impressed that some of the most unusually healthy patients with the most indolent lung cancer I’ve ever seen are the ones who jump on a plane to Dallas to enroll on these trials led by Dr. John Nemunaitis at Baylor, who runs multiple vaccine-based trials — because of this, I think it will be very important to rule out selection bias of remarkably unusual patients as a reason that these patients did so much better than we expect to see in large trials of previously treated patients. in fact, the reported “tumor burden” (basically, the size of measured disease) was very low in their responding patients, so I wonder if many of these patients had far less aggressive disease than a more average population that would be reflected in results from larger trials. But perhaps suggesting a real benefit, responses on scans were correlated with better immune responses in patients on the trial.
On further review of the data, results seemed best among patients who had achieved stable disease or better on prior treatment (although it’s pretty much this way with every treatment — “responders respond“). Combined with the rest of the findings, this led to development of a phase III clinical trial with up to 700 patients who have received prior chemo for advanced NSCLC and shown stable disease or better after first line treatment to then be randomized to active vaccine or placebo (50% chance of each). This trial is just getting started now (more information here). While I do think this approach is promising enough to be worth testing in a larger setting, and I think the early results desperately need to be compared to a control arm in a randomized trial to separate true treatment effect from the strong possibility of selection bias, I think this will be a very hard trial to complete. First, placebo-controlled trials are the easiest to interpret and “purest” in design, but the fact is that many patients are reluctant to enroll, especially American patients, because they express an extreme aversion to being potentially randomized to a placebo. While it is ethically permissible to offer a placebo in a setting of “maintenance” where it’s not yet a standard to do anything but follow a patient who has shown a response or stable disease after first line therapy, I’ve written that I believe that the evidence is compelling enough that I would now be inclined to recommend an immediate transition from first line to second line treatment (see prior post). Finally, I think it will be challenging to get enough patients enrolled who have shown stable disease or responded to first line therapy and are now off of treatment, not yet on second line therapy, and not yet progressing. That’s really threading a needle, and trials that have had such a narrow window for enrollment have had a very difficult time accruing patients.
That said, this work looks intriguing enough that I really hope I’m wrong, that many patients are interested in enrolling and the trial accrues quickly. I’d be interested to hear whether people here would be willing to enroll on such a trial, or whether they would be unwilling to subject themselves to be randomized to a placebo in the maintenance setting after first line treatment.