Top, from left: Vivian Liu, Steve Carr, Vamsi Mootha.
Bottom, from left: Romain Guièze, Cathy Wu, John Doench
In their own words, Dana-Farber and Broad scientists describe how their collaboration revealed key changes in metabolism in drug-resistant leukemia cells
The targeted cancer drug venetoclax, first approved by the FDA in 2016, brought new hope for patients with blood cancers such as chronic lymphocytic leukemia (CLL). But that enthusiasm began fading as the drug stopped working for some patients who saw their disease return and their treatment options dwindle.
For Catherine Wu, a physician-scientist at the Broad Institute of MIT and Harvard and Dana-Farber Cancer Institute (DFCI), the cancer’s ability to evade this drug was a call to action.
With her colleagues at DFCI and the Broad, Wu set out to analyze how CLL cells become resistant to venetoclax. The drug targets the molecular roots of CLL and other cancers by inhibiting the BCL2 protein, which is abundant in CLL cells and protects the cell by preventing the formation of holes in the membranes of cellular organelles called mitochondria.
By joining forces with experts at the Broad in the fields of proteomics, metabolism, and genetic perturbation, Wu and her team found that drug-resistant CLL cells survive by overexpressing the MCL1 protein, and also by altering how mitochondria use energy. These insights, published recently in Cancer Cell, suggest new combinations of therapeutics that could one day treat cancer more effectively.
Here, the scientists describe their journey to discovery in their own words.
Cathy Wu, Institute Member in the Broad Institute Cancer Program, Associate Professor of Medicine at Harvard Medical School, Staff Physician at DFCI and Brigham and Women’s Hospital: In the past few years, we’ve seen the treatment landscape for CLL change before our eyes. Until a couple of years ago, the standard treatment was chemotherapy, but now people are getting more efficient, less toxic medication that targets CLL’s molecular roots. The approval of venetoclax has been one of the most encouraging developments in this disease, and being able to witness how effective it is has been really exciting. However, resistance is still a problem.
Romain Guièze, Hematologist at Centre Hospitalier Universitaire de Clermont-Ferrand, former researcher in the Wu lab: As soon as venetoclax was approved, we started anticipating how the cancer might evade treatment. As with other targeted drugs, we’ve seen a good proportion of patients who relapse on venetoclax, despite an initial strong response, and these patients have a poor outcome and limited options. It is crucial that we understand the rules underlying that process of resistance in order to maximize the benefit from these novel agents.
Vivian Liu, Internal Medicine Resident at University of Washington Affiliate Hospital, former researcher in the Wu lab: I’ve seen a few patients on venetoclax or another targeted therapy and I’m hopeful for a future in which more and more patients can benefit from these new drugs. In order for venetoclax to achieve its full potential, we must find the relevant mechanisms underlying resistance. Systematic, multi-faceted approaches are critical to uncovering unexpected pathways of resistance, which in turn shed light on the biology of both the disease and the drug.
Catherine Wu: Our first step was to work with Gaddy Getz in the Broad’s Cancer Program to sequence DNA from tumor cells taken from patients before venetoclax treatment, and after relapsing. We saw that CLL cells from a single patient can carry different mutations, and that cells with some of these mutations become more abundant after treatment, but we didn’t see anything obvious to explain resistance. It was clear that we needed to look in an unbiased and systematic way at how different mutations and genes impact the function of cancer cells, to reveal how a heterogeneous population of tumor cells can escape drug treatment.
The scientists next reached out to John Doench and his team in the Broad’s Genetic Perturbation Platform, to systematically alter genes in the cancer cells and see if the altered cells survive drug treatment. With John and his team, they brainstormed how to design the best screen to run, determine the right experimental setting, construct the screen, and maximize the information to be gained from it.
John Doench, Associate Director of the Genetic Perturbation Platform at the Broad: Perturbing a biological system in multiple ways can provide a much clearer picture than only one. Performing single perturbations in isolation is like five blindfolded people all trying to identify an elephant by touching different parts, and all misidentifying it. (It's a snake! It's a spear! It's a wall!) Only by sharing and integrating information from different perspectives can we surmise that it’s an elephant.
My colleague Federica Piccioni and I suggested that Cathy and her team would get the most insight by not only turning genes off, but also seeing what happens when they’re turned on, known as gene “overexpression.” Once we’ve done one type of perturbation [for example, open reading frames (ORFs), knockouts with CRISPR, overexpression with CRISPRa, RNA interference], and we’ve built a relevant model system and coupled it to an informative assay, why not screen it with as many things as possible.
Doench and Piccioni also advised that the team examine not just the top hits from the screens, but all of the data, to generate new hypotheses. For help analyzing the screening data, the researchers consulted with Cory Johannessen, now a senior investigator at Novartis Institutes for BioMedical Research who at the time led a Broad team looking for drug-resistance mechanisms and effective drug combinations.
Catherine Wu: With Cory’s help, we had the opportunity to think deeply about the hits from our functional screens, and whether each might underlie drug resistance. There were lots of unexpected and unknown hits, and that’s where the power of our environment at the Broad and Dana-Farber really came to bear. The data showed that resistant cells make too much of a protein called MCL1. Our colleague Guo Wei [now a principal scientist at Sanofi who was then a researcher at the Broad] had been developing an MCL1 inhibitor, so we connected with Guo to test and confirm our hypothesis that MCL1 overexpression drives resistance.
The team also studied a venetoclax-resistant cell line using transcriptomic methods in the Wu lab and proteomic approaches in the Broad’s Proteomics Platform, to observe which genes were being read and made into protein more often in resistant cells than in drug-sensitive ones.
Steve Carr, Director of the Proteomics Platform at the Broad: Namrata Udeshi led our platform’s mass spectrometry-based proteomics analysis of a venetoclax-resistant cell line. Her analysis revealed dysregulation of proteins in the resistant line originating from genes critical to cellular metabolism, cell cycle, B-cell biology, and autophagy. These findings were also reflected in the RNA analyses conducted in Cathy’s lab. In light of the genetic screening hits that were also involved in cellular energy metabolism, it became clear that dysregulation was probably related to altered mitochondrial biology, so I connected Cathy to our colleague, Vamsi Mootha.
Vamsi Mootha, Co-director of the Metabolism Program and an Institute Member at the Broad, Professor of Systems Biology and Medicine at Harvard Medical School, Professor of Medicine at Massachusetts General Hospital: After brainstorming about next steps, Alexis Jourdain, a postdoctoral fellow in my lab based at the Broad Institute, examined the effects of venetoclax on mitochondrial physiology. He discovered that venetoclax treatment led to a dramatic reduction in mitochondrial respiration, which is central to mitochondrial energy metabolism. Importantly, he observed that venotoclax-resistant cell lines were immune to this inhibition and could maintain respiration even in the presence of a high dose of venetoclax.
It’s long been appreciated that mitochondria play central roles in energy metabolism as well as in programmed cell death (apoptosis) — what’s neat about this paper is that it connects these two functions in a way that may be relevant for disease and therapy.
The team went on to confirm their results using patient tissue samples. Now, Wu and her team are exploring whether the findings might apply to other diseases that are treated using venetoclax, or whether combinations of drugs could help overcome resistance.
Catherine Wu: In science, we don’t always have the resources to deeply explore results from open-ended studies, so those efforts remain open-ended and you don’t know what the results mean. In this case, it was beautiful and satisfying to start with an open-ended question and then have the resources and expertise at the Broad and Dana-Farber to follow it through to the end, and hopefully we can one day bring it back to the clinic to benefit patients.