An important step for a rare brain tumor - Results of a Phase II trial of BRAF/MEK inhibition in newly diagnosed papillary craniopharyngiomas

Very excited about the publication today of the results of Brastianos et al., Alliance A071601 (Phase II trial of BRAF/MEK inhibition in newly diagnosed papillary craniopharyngiomas) in the New England Journal of Medicine.

Some reflections on the decade of work leading to this important milestone.

First, this feels like a triumph in the realm of rare diseases, particularly in the world of rare brain tumors. There have been so few successes in the precision medicine for brain tumors, where the identification of a specific genetic mutation enables a targeted therapy that achieves a dramatic clinical result. This is among the most striking examples to date. This remarkable clinical achievement is the result of a major effort led by Priscilla Brastianos, MD, Fred Barker II, MD, Eva Galanis, MD, and many colleagues at MGH, DFCI/BWH, Mayo Clinic, BCH, Alliance for Clinical Trials in Oncology, and more around the country. Quite a remarkable collaboration.

A short primer

There are two types of craniopharyngiomas, both with names that are a mouthful – 1. Adamantinomatous craniopharyngioma (ACP for short) which occur in both children and adults and 2. Papillary craniopharyngioma (PCP for short) which mainly occur in adults, but with exceptionally rare cases arising in children. Both of these are classified as very rare brain tumors. Annually, about 500 individuals develop ACP and about 125 develop PCP in the United States. These tumors arise in the area of the pituitary gland right next to the optic nerves, so they are challenging to treat, and many patients unfortunately develop serious lifelong endocrine and neurological complications that can be difficult to manage. The story of a young girl’s struggles with ACP from 2013 still breaks my heart every time I read it.

In 2014, in Brastianos et al. (Nature Genetics) we reported that nearly all ACP have mutations in CTNNB1 (the gene that encodes beta-catenin) whereas remarkably nearly all PCP have BRAF V600E mutations. This striking difference in the genetic driver has now allowed us to classify ACP and PCP as two clearly different tumor types - now fully separated into two categories in the World Health Organization Classification of Tumors of the Central Nervous System.

This work was followed soon after by a heroic N of 1 study (Brastianos et al. JNCI) showing that a patient with recurrent BRAF V600E mutant PCP had a remarkable 85% reduction in tumor size after only one month of targeted BRAF/MEK inhibitor therapy.?

That set the stage for the striking findings reported today from Alliance A071601 in NEJM.

Reflecting on a few important points, lessons, and observations from the long arc of this work:

Serendipity is a Critical Part of Science.

In the 2014 study we set out to find genetic drivers of ACP in pediatric samples from Boston Children’s Hospital in collaboration with Mark Kieran, MD, Peter Manley, MD, and colleagues. Being the most common type of craniopharyngioma, our team’s primary goal was to understand if we could find information in the genetics of ACP that would lead to a new treatment approach, beyond surgery and radiation. The cohort of ACP samples we collected just happened to also have one PCP sample, unintentionally (recall – PCP is very rare in children). This is a good example of why it is important to be prepared and alert for the unexpected.

Commitment to Research is Key.

That one, single, individual PCP sample was available to us due to a family’s commitment to supporting research, and their decision to consent to the use of the surgically resected tissue. I still find this goodwill and the good fortune we had truly amazing. Tissue samples of rare tumors can be particularly precious – we do not have any cell lines or animal models of PCP. All of the progress that was made was based on tissue from our pathology archives and tissue bank, and clinical testing performed on patients.

Academic Science and Academic Pathology are Very Important.

In early 2013, due to the commitment of our academic translational research community we had access to early targeted sequencing technologies at both DFCI’s Center for Cancer Genomics (formerly CCGD – thank you Aaron Thorner, Laura MacConaill, Bill Hahn, Paul Van Hummelen, and Matthew Meyerson!) and MGH’s Translational Research Laboratory (set up by John Iafrate, MD, and by my wife Dora Dias-Santagata, PhD, … always nice to have close collaborators in high places!!??). We had access to methods – OncoMap and SNaPshot and other sequencing technologies – some of which would now be considered somewhat medieval but which few others had readily available at that time. This actually reminds me of the early access we currently have for spatial imaging tools and how important it is that academic medical centers continue to invest in leading in the development of cutting edge methods and supporting their faculty so that they have the time and resources to use these tools in creative and successful ways.

Robust sequencings technologies are ubiquitous now – but certainly not then. Also, we benefited so much from the generosity of academic pathology collaborators in Boston (MGH, BWH/DFCI, BCH) and around the country (particularly UPenn, CHOP, Hopkins) that allowed us to quickly collect additional samples of PCP (39 samples in total!). We used the sequencing technologies at DFCI/BWH and MGH to show that essentially all PCP had BRAF V600E mutations (~95%), with additional detailed and important genomic analysis performed in collaboration with colleagues at the Broad (a lot of diligent work by Amaro Taylor-Weiner who was in Gad Getz’s team). All of this is fairly straightforward to do nowadays, not so easy then. Having a number of ACP samples available for comparison helped a lot too (many samples were received through a generous collaboration with Hala Taha, MD, at the National Cancer Institute, Cairo University, Cairo, Egypt). On top of that, we had the support of multiple clinical research coordinators at all levels in Neuro-Oncology at DFCI (thank you Will Pisano, Jack Geduldig, and others as part of Keith Ligon’s team!), MGH, and Alliance, and the staff in the research and clinical labs. To me, this represents academic medicine at its best – open collaborative sharing of critical materials and ideas, that permitted us to pursue a potentially important finding.

Also – I would be remiss if I did not give a loud shout out to us pathologists in the field of Academic Pathology. Classifying tumors is what we all do, all the time (largely in obscurity) – my colleagues across pathology have vast amounts of practical knowledge about tumors that has been developed over decades of close inspection of tissues. The contributions of pathologist to tissue-based research are fundamental and vast. Louis Pasteur said, "In the fields of observation, chance favors only the prepared mind." Pathologists certainly have those prepared minds. 100%.

Small Foundations Matter So Much.

The National Institutes of Health (NIH) and the National Cancer Institute (NCI) are the bedrock of much of our biomedical research (certainly in my lab), providing absolutely critical funding for us to push the frontiers of basic and translational research. We are very lucky indeed. However, small focused translational projects such as on tumors like ACP and PCP are often difficult to fund – the clinical need is not as high obviously for these rare tumors as it is for very common tumor types. So study sections reviewing grants are less likely to be enthusiastic about such proposals. Also, the type of work that needs to be done is generally not considered ‘hypothesis driven’ mechanistic science, which is often prioritized over nuts-and-bolts data collection and practical steps in medicine - that just happens to be less attractive to fund. I personally have never directly held funding to work on craniopharyngioma.

This is where small and determined foundations that support cancer research (and their tireless and dedicated supporters and fundraisers) play a massive role. The early work that led us to find BRAF V600E mutations in PCP was largely supported by foundations that funded our colleagues in pediatric Neuro-Oncology – the Jared Branfman Sunflowers for Life Fund for Pediatric Brain and Spinal Cancer Research, the Pediatric Low-Grade Astrocytoma (PLGA) Program, Pedals for Pediatrics and the Clark Family, the Stahl Family Charitable Foundation, the Stop&Shop Pediatric Brain Tumor Program, the Pediatric Brain Tumor Clinical and Research Fund, as well as other foundations: the V Foundation and the American Brain Tumor Association. Lots of bike rides, drink stands, and bake sales, and more. If you ever wonder if any of that helps, these contributions add up to important steps in medicine. Without foundations, big and small, our work would not have been done. Also, it is important to note that sometimes, foundations fund ‘Disease X’ but we make breakthroughs in ‘Disease Y.’ That’s just the way it works. But no worries – then someone else funds research in ‘Disease Z’ which leads to breakthroughs in ‘Disease X.’?It is all interconnected … step by step we make progress.

It is Important to Focus Resources on Reuse and Repurposing of Innovative Therapies.

Finding treatments for rare cancers is very hard for many reasons. The staggering cost of developing drugs combined with limited market potential for the small population of patients with rare tumors can make it difficult for pharmaceutical companies to justify investment. The success in BRAF V600E mutant PCP was based on foundational work done by the melanoma community, based on the frequent (~50%) presence of BRAF V600E mutations in melanoma. The FDA approvals made in 2011 to 2013 for drugs that target the BRAF-MEK pathway for use in treating melanoma patients provided a new therapeutic option for patients with BRAF-mutant tumors of all different types, including other rare BRAF mutant diseases such as Erdheim-Chester and Langerhans Cell Histiocytosis, as well as now PCP. In fact, we subsequently reported in 2016 the successful use of a BRAF inhibitor for treating a patient with another rare brain tumor, a rare type of glioma called pleomorphic xanthoastrocytoma (PXA; Lee EQ, et al. JCO, 2016).

Finding ways of taking treatments that are successfully developed for common tumor types (breast, colon, melanoma, lung, etc) and reusing them in rare tumors seems to be the most effective way forward – this is a main focus of a number of our translational research efforts. Nonetheless, it still took a decade from the time of finding mutations to reporting a successful phase 2 trial. And moreover, even with good clinical rationale, it is still quite difficult to secure drugs for brain tumor trials, as well as the funding to test FDA approved therapies. This remains a very real and major barrier to progress that we struggle with constantly.

Tenacity of Neuro-Oncologists to Leverage the National Clinical Trial Infrastructure.

Priscilla Brastianos, MD, Mass General Cancer Center, has led the charge, from the early days of characterizing the frequency of the mutations, through the first proof-of-concept clinical studies, to organizing and conducting the Alliance trial A071601 as the Study Chair (along with Study Co-chairs Eva Galanis, MD and Frederick Barker, MD) and working closely with so many highly engaged colleagues at the NCI’s National Clinical Trials Network and the Alliance – the national organizations and clinicians that coordinate and support cancer clinical trials in our country. There are very people that I know of that would have had the willpower to bring all of this together.?Patients with PCP (and brain tumors in general) could have no better champion than Dr. Brastianos!

Unfinished work.

More work remains to be done and we are at it. There are many questions remaining for patients with PCP – can we find additional therapies for patients that can’t tolerate extended treatment on current regimens, how long can patients stay on combination therapy, what is the role for monotherapy, etc. There is more work to do.

ACP remains a difficult challenge. A great frustration is that we still do not have effective medical treatments for patients with this disease. I feel very bad about this. Our group, and others around the world are focused on changing that. But rare tumor research is slow and still hard to support. So, our approach is to include rare tumor samples in studies where the main goal is to focus on more common tumor types … so, it is not uncommon in our papers that we report the results of studying a biomarker or a pathway in glioblastoma (the most aggressive brain tumor type) but also report data on a handful of ACP and PCP specimens and other rare tumor types like diffuse intrinsic pontine gliomas (DIPG).

Example - we included data on 22 ACP, 16 PCP, and 11 diffuse midline gliomas in Figure 4 of this study. Planning to slip in similar data on these rare tumor types in another study we currently have under review.

This is our current approach to try and continue to make progress in rare brain tumors.

The success of Alliance A071601 for PCP makes me hopefully that progress is forthcoming for ACP and for that matter other brain tumors.

Ok, now back to work …?