Independent ecancer blogger Dr Bishal Gyawali rounds up the latest publications in oncology.

Nobody knew that cancer could be so complicated

Two important papers from the TRACERx consortium made big news in lung cancer in May: one published in NEJM and the other in Nature. The investigators interrogated 327 tumour regions on 100 early stage NSCLC tumours that had been resected before systemic therapy and found widespread intratumour heterogeneity. The driver mutations in EGFR, MET, BRAF and TP53 were clonal while other driver mutations that occurred later in evolution were heterogeneous and occurred commonly in PIK3CA, NF1 and genes involved in chromatin modification and DNA damage response and repair. This heterogeneity mediated through elevated copy number alterations and this was associated with nearly 5 times increased risk of recurrence or death which was significant. The study published in Nature shows that the evolutionary events of lung cancer including chemo resistance, recurrence and emerging subclones seeding metastatic sites could be depicted non-invasively by ctDNA analysis obviating the need for tumour biopsy and providing further insights into the evolutionary history. Both these studies have huge potential for future therapeutic application. For today, if anything, the complex heterogeneity suggests that precision medicine based on matching the name of a drug to the name of a mutation might not work as well as we have imagined them to. Future therapies should focus on ways to minimize copy number alterations or intratumour heterogeneity, if that is possible at all! Indeed, “It's an unbelievably complex subject. Nobody knew that cancer biology could be so complicated."

When improved PFS may not mean benefit

There are certain instances in oncology where the improvement in PFS alone may be deceptive and shouldn’t be the basis for approving therapies. We discuss these issues in detail in a recent paper published in Nature Reviews Clinical Oncology.  One such instance involves giving maintenance therapy. Improvement in PFS is obvious with maintenance therapy because we are giving an active drug; but if it doesn’t improve survival, that would only mean increased toxicity. The second scenario involves combining drugs that are already approved for sequential use. This is very topical and important because the FDA has recently approved pembrolizumab  chemotherapy combo for first line use in non- small cell lung cancer. This approval is based on improved PFS with chemo plus pembrolizumab versus pembrolizumab alone in a small phase 1/2 open-label study. This is baffling for many reasons: firstly, we don’t know if the PFS (much less OS!) of pembrolizumab plus chemotherapy is better than pembrolizumab followed by chemotherapy. As we describe in our paper if treatment A has a PFS of 5 months in 1st line and treatment B has a PFS of 4 months in 2nd line, a trial that finds A plus B has a better PFS of 7 months in 1st line will be misleading because the 7 months of PFS would in fact be worse than the PFS of A followed by B (9 months). Secondly, this was not a placebo controlled study. In studies that are not blind, PFS is a very unreliable endpoint. Thirdly, this was a phase 1/2 study. As I discuss below with selumetinib, benefits in phase 1/2 are exaggerated and get lost in phase 3.

The PFS has improved but will patients live longer?

Another example where caution should be exercised in interpreting the PFS data is the J-ALEX trial. In this trial, alectinib used first line among Japanese patients with ALK positive lung cancer had significantly improved PFS versus the standard of care crizotinib. Full marks for comparing alectinib against the standard of care crizotinib; unlike for example ceritinib that randomised the patients in the control group to chemotherapy instead! However, alectinib is approved and active in second-line, i.e. among patients who have progressed on first-line crizotinib but the reverse is not known. Hence, using alectinib upfront might mean depriving patients of crizotinib. A better PFS with alectinib versus crizotinib is not sufficient to make proper judgment: the PFS with alectinib should be better than crizotinib followed by alectinib in sequence. Preferably, OS data would alleviate all concerns.

SELECT phase 3 for drug approval

Phase 2 trials, even if randomized and placebo-controlled, may lead to misleading results. Thus, a phase 3 RCT is important even if a phase 2 shows efficacy. At a time when the regulatory agencies have started approving drugs based on response rates alone, you might call me crazy to ask for phase 3 despite phase 2 showing positive signals but take a look at selumetinib in lung cancer. A previous placebo-controlled multicenter RCT had showed significant improvement in PFS with selumetinib versus placebo added to docetaxel (median 5.3 v 2.1 months) with good hazard ratio (0.58, p = 0.014). I don’t know why they didn’t file for accelerated approval based on this data alone! Anyway, we now have a phase 3 which showed no benefit in PFS and OS. This is a nice example that exemplifies the need for a phase 3 RCT. But we are so lowering the bar for approval that I find it funny to be asking for phase 3 RCTs when drugs are getting approved based on response rates alone!

Speaking of accelerated approvals, another study in JAMA concludes that “biologics, psychiatric therapeutics, and accelerated and near–regulatory deadline approval were statistically significantly associated with higher rates of events, highlighting the need for continuous monitoring of the safety of novel therapeutics throughout their life cycle.” More than 21% of the drugs studied in this analysis are drugs used in the treatment of cancer and hematological diseases.

T-DM1 in breast cancer

TH3RESA final survival results were published in MAY and together with EMILIA now establish the benefit of T-DM1 in the treatment of metastatic HER2 positive breast cancer previously treated with trastuzumab containing regimen. However, T-DM1 shouldn’t be used in first-line or in combination with pertuzumab, given the negative results with MARIANNE.

Screening: enter as a person, exit as a patient

The U.S. Preventive Services Task Force has recommended against screening for thyroid cancer in asymptomatic adults. This should be a no brainer and even without any guidelines, no doctor should have recommended thyroid cancer screening especially after the reports on the tsunami of overdiagnosis of thyroid cancer surfaced last year (especially from Korea). Last August, I wrote about thyroid cancer screening in my blog which I still find very relevant: “With increasing precision in diagnosis, we are faced with increasing uncertainties in making treatment decisions. For instance, we are now detecting an alarmingly high number of thyroid cancer cases with precision, but are uncertain about how best to manage them. There is no evidence that detecting these thyroid cancers early with ultrasound scan would change the outcomes. Notwithstanding, we have been performing a huge number of scans, detecting a large number of early thyroid cancers and wondering what to do with them. But the problem is that we don’t just wait and wonder! We want to do something with that tiny thyroid mass now that we have detected it! Doing satisfies the vanity of us being doctors. As a result, many of us end up harming the patients by offering surgeries and invasive treatments; although our intentions were noble. Thus, improved precision is not always a good thing. Why would you want to find something if you don’t know what to do with it?”. Medicine is supposed to have its foundation on “Primum Non Nocere” but modern oncology is more about converting people into patients and making their lives miserable either way: once you are told that you have a thyroid mass, you are in a lose-lose situation. If you don’t undertake surgery, you will live your whole life under the fear of what if it were cancer. If you undergo surgery, you’ll not only suffer unnecessary surgery for something that was never going to harm you but you’ll also need to depend on hormone supplements all your life. In one of the amazing editorials I have read recently, Dr. Welch walks you through the whole (non) sense of “Cancer Screening, Overdiagnosis, and Regulatory Capture”.

The purpose of drug repurposing

I admit I am a big fan of drug repurposing because it represents an untapped reservoir of cheap drugs that can be used for treating cancer. Given how marginal the modern cancer drugs are, I am actually optimistic that the repurposed drugs could easily meet these modern drugs. Plus, they are relatively safe with years of safety data available. Drug repurposing would be an important measure to address financial toxicity of cancer treatment. That’s why I have recently decided to support the not-for-profit Anticancer Fund, which focuses on identifying and trialing such repurposed drugs, as a medical consultant. However, in another sad story of drug repurposing, celecoxib failed to improve outcomes when added to chemotherapy for lung cancer in a phase 3 RCT.  There are a few positive takeaways nevertheless: first, well-conducted RCTs of repurposed drugs are possible; second, cooperative groups and academia should take initiatives for such trials that are unlikely to be conducted by the industry; third, a well conducted RCT generates strong evidence and can be conducted with the same funding that would otherwise be spent on hundreds of observational studies testing the same question.

Speaking of cheaper interventions, it is pertinent to introduce this study here that shows blended cognitive behavioral therapy is helpful to prevent the fear of recurrence among cancer survivors curatively treated for breast, prostate, or colorectal cancer. Now this is a worthy endpoint that actually matters and a cheap intervention. Will this be widely adopted? Chances are slim because this intervention doesn’t have a “mab” added to its end and doesn’t cost $10k a month.

Let me take a selfie

The new cancer drugs defy all logic and keep getting expensive despite only marginally improving outcomes. In a new open-access cancer policy paper I coauthored with Prof. Richard Sullivan, we try to investigate “the landscape for cancer medicines that is becoming increasingly confused by propaganda, a lack of normative principles upon which to base policy and a scientific paradigm that, because of its inherent complexity, is increasingly becoming a ‘black box’ to those outside the scientific community.” One such paradox that we highlight includes “it seems a bit naïve that the HICs spend millions in basic research identifying a target (public research) but then allow private (and limited to shareholder) profit. The most important equation dictating the price of new drugs — the clinical trials leading to approval— is completely out of public control, and left solely to the private sector.” Finally, we propose some solutions about the way forward.

Bishal Gyawali, MD is a postgraduate trainee in medical oncology at the Graduate School of Medicine, Nagoya University, Japan, where he is also a PhD candidate under the Japanese government scholarship. He is also a medical consultant at Anticancer Fund, a not-for-profit organization based in Belgium as well as holds an affiliation at Institute of Cancer Policy, UK. His areas of clinical and research interests include evidence-based oncology practice, global oncology, cancer policy, cancer management in resource-limited settings, financial toxicities of cancer treatment, clinical trial methods and supportive treatment of cancer. He has no conflicts of interest to declare. Dr Gyawali tweets at @oncology_bg. Read his previous blog posts here.

Access our recently launched e-learning on drug repurposing here.