View an archive of research highlights and featured publications from the Center for Cancer Research.


Immunity to Commensal Virome Blocks Cancer

Demehri lab
Right to left: Jonathan Messerschmidt, Shawn Demehri, Tiancheng Li

In an October 2019 paper published in Nature, Strickley, Messerschmidt et al. investigated the role commensal human papillomaviruses (HPVs) play in the development of skin cancer. Using a novel skin colonization model, they show that adaptive immunity against papillomavirus infection is able to protect the host from developing cancer. In stark contrast to the established “hit and run hypothesis,” which postulates that skin HPVs initiate cancer and are later lost, this work elucidates the significant and previously unrecognized beneficial contributions of skin-resident HPVs. Considering the >100-fold increase in skin cancer risk among immunosuppressed patients, these findings potentiate novel therapeutic interventions to prevent the development of skin cancer in this high-risk population. Furthermore, these results establish a new field of investigation into the beneficial contributions of viruses that live in our skin and other organs. View the press release.

Visit the Demehri lab.

CAR-T cells secreting BiTEs circumvent antigen escape without detectable toxicity.

Marcela Maus (left), Bryan Choi (center), and Xiaoling Yu (right)
Marcela Maus (left), Bryan Choi (center) and Xiaoling Yu (right)

Building on their research showing that an exciting new form of immunotherapy for cancer has activity in patients with glioblastoma, the most common and most deadly form of brain cancer, Mass General investigators have created a new method that could make immune therapy more effective again brain tumors and expand its use against other types of solid tumors. Their study is published in the July 2019 issue of Nature Biotechnology. View the press release.


An Integrative Model of Cellular States, Plasticity, and Genetics for Glioblastoma.

Mario Suva (left) with Toshiro Hara (right) and Marni Shore (center)

The researchers profiled gene expression in more than 24,000 tumor cells from 20 adult and eight pediatric glioblastoma patients and also analyzed glioblastoma models in the lab. They found four glioblastoma cellular states, which each has a unique gene expression program and together help account for the large variation in the disease. The scientists then used the single-cell data to reanalyze glioblastoma data from The Cancer Genome Atlas, and identified genetic alterations associated with each of the four states. The results, published in Cell, also show that the cells are remarkably flexible or plastic — that is, they can switch between the four states. This shape-shifting ability could help explain why these cancer cells are so difficult to kill with drugs and help inform the development of better therapies for glioblastoma. The research was led by co-first authors Cyril Neftel, Department of Pathology at Massachusetts General Hospital (MGH) and Broad Institute; Julie Laffy, Weizmann Institute of Science; Mariella Filbin, Department of Pathology at MGH, Broad Institute, and Dana-Farber Institute; Toshiro Hara, Salk Institute for Biological Studies; and co-senior authors Itay Tirosh, Weizmann Institute of Science, and Mario Suvà, Department of Pathology at MGH and Broad Institute. View the press release.

Visualizing Engrafted Human Cancer and Therapy Responses in Immunodeficient Zebrafish.

Dave Langenau (left) with Chuan Yan (right)

In a June 2019 paper published in Cell, Yan, Langenau and coworkers show that they can engraft a wide array of human tumors in optically clear zebrafish. The advantages of this system for imaging mean that it is readily possible to visualize dynamic changes in single engrafted cells. The authors have used their zebrafish models to show the preclinical efficacy of a drug combination of the PARP inhibitor Olaparib and the DNA-damaging agent Temozolomide for rhabdomyosarcoma. The ability to grow patient-derived tumors in zebrafish will likely open many new avenues for personalized medicine.

Passenger hotspot mutations in cancer driven by APOBEC3A and mesoscale genomic features.

Lawrence Lab
Left to right: Adam Langenbucher, Dr. Michael Lawrence, Dr. Rémi Buisson, Dr. Cyril Benes, Dr. Lee Zou.

The DNA of cancer cells contains many mutations. A few of these mutations are called "driver mutations" because they were responsible for changing the behavior of the cells to make them multiply out of control and form a tumor. But the vast majority of mutations in the cell had no effect and are just "along for the ride" - these are called "passenger mutations". Telling driver mutations apart from the far more numerous passenger mutations can be very challenging. One commonly used approach is to look for exactly the same mutation occurring in many different patients' cancers. This "recurrence" approach has been very successful over the past decade at identifying cancer driver genes and mutations. However, it assumes that every recurrent mutation ("mutation hotspot") is a driver hotspot. In this 2019 paper published in Science, MGH researchers Dr. Lee Zou and Dr. Michael Lawrence took a new critical look at this assumption. Combining the power of biochemistry and bioinformatics, they and their co-authors showed that some hotspots are actually "passenger hotspots," a term that would be seen as contradictory given the current thinking of the field. These passenger hotspots result from the very specific preferences of APOBEC3A, a DNA-mutating enzyme that is often upregulated in cancer cells. They showed that APOBEC3A has a very strong preference for mutating C nucleotides exposed in a short loop at the end of a DNA "hairpin". DNA hairpins form when nearby palindromic DNA sequences in one DNA strand pair to each other instead of their complementary partner on the opposite DNA strand. This generates a perfect substrate for APOBEC3A to attack, and leads to the formation of passenger hotspots, which are recurrently mutated in many patients' cancers, despite contributing nothing to the development or fitness of the cancers. These results show that some previously nominated cancer drivers (e.g. a hotspot in the gene MB21D2) are actually passenger hotspots, and would likely not be worthwhile targets for further study. Incorporating these insights into driver discovery algorithms will improve our ability to detect the true drivers of cancer.

Transcriptome-wide off-target RNA editing induced by CRISPR-guided DNA base editors.

L to R: Keith Joung, MD, PhD, Caleb Lareau, Julian Grunewald, MD, Ronghao Zhou, Sowmya Iyer, PhD, and Sara Garcia, PhD

In a May 2019 paper published in Nature, Grünewald et al investigated the effects of cytosine base editors for RNA and DNA editing. They show that targeting of rat APOBEC1 can cause extensive transcriptome-wide deamination of RNA cytosines in human cells, inducing tens of thousands of C-to-U edits. They engineered two variants bearing mutations in rat APOBEC1 that substantially decreased the number of RNA edits (by more than 390-fold and more than 3,800-fold) in human cells. These variants also showed more precise on-target DNA editing than the wild-type protein and, for most guide RNAs tested, no substantial reduction in editing efficiency. These results, together with experiments with an adenine base editor, have important implications for the use of base editors in both research and clinical settings.


Developmental and oncogenic programs in H3K27M gliomas dissected by single-cell RNA-seq.

In this April 2018 paper published in Science, a group of researchers from the MGH and the Broad Institute used single-cell RNA-seq to define the cellular context of diffuse midline gliomas with histone H3 lysine27-to-methionine (H3K27M) mutations. These uniformly fatal malignancies arise primarily in the midline of the central nervous system of young children, suggesting a particular cell type present at this location and developmental stage is susceptible to transformation by H3K27M. By sequencing the RNA from 3321 single cells from 6 tumors, the authors found that H3K27M tumors largely consist of cells that resemble oligodendrocyte precursor cells (OPC) and drive the growth of the tumor, whereas more differentiated cells were in the minority and lacked proliferative potential. The authors suggest that targeting the OPC marker PDGFRA or triggering cells to differentiate might allow therapeutic intervention in these as yet incurable malignancies. For more information, click here.

This work was led by first authors Mariella Filbin (pictured), Itay Tirosh and Volker Hovestadt (pictured). It was made possible by a close collaboration between the laboratory of Mario Suva’s (pictured), Brad Bernstein and Aviv Regev.

A mitosis-specific and R loop-driven ATR pathway promotes faithful chromosome segregation.

ATR is a master kinase that regulates the DNA damage response, and it is particularly important during DNA replication to keep the genome stable. Although loss of ATR loss was known to cause problems in mitosis, these effects were thought to be indirect. In a January 2018 paper published in Science, Lilian Kabeche and colleagues in Lee Zou’s lab (pictured) report their discovery of a surprising function of ATR in mitosis. In early mitosis ATR is localized to and activated at centromeres. The activation of ATR at centromeres requires R loops, a transcription intermediate containing DNA:RNA hybrids and single-stranded DNA. Active ATR at centromeres promotes accurate segregation of whole chromosomes by stimulating the Aurora B kinase, ensuring that microtubules are attached to chromosomes properly. This study uncovers the first R loop-driven signaling cascade on chromosomes, and a new ATR pathway that specifically operates in mitosis, revealing an unexpected dimension of cell cycle regulation.


TOX regulates growth, DNA repair, and genomic instability in T-cell Acute Lymphoblastic Leukemia.

In this October 2017 paper in Cancer Discovery, a research team led by David Langenau that included Marina Theodorou (pictured) identified thymocyte selection-associated high mobility box protein (TOX) protein as a major oncogenic driver responsible for transforming normal T cells into aggressive leukemia. TOX is also required for leukemia cell growth and thus represents a new, previously unexplored therapeutic target in this disease. Importantly, the authors found that TOX is expressed in 95% of human T cell acute lymphoblastic leukemia suggesting TOX is a major oncogenic driver in a vast majority of patients with this disease. Using a wide array of functional analyses, TOX was found to bind directly to Ku70/80 and to suppress recruitment of this complex to DNA breaks, thus inhibiting non-homologous end joining repair, which is known to cause genomic instability. Collectively, this work has uncovered important roles for TOX in regulating elevating genomic instability during leukemia initiation and sustaining leukemic cell proliferation following transformation. 

getz labA mutational signature reveals alterations underlying deficient homologous recombination repair in breast cancer.

Breast cancer cells with defects in two genes involved in homologous recombination, BRCA1 and BRCA2, have a specific pattern of mutations in their genomes. This pattern, known as Signature 3, is also found in some patients without mutations in BRCA1/2. In the August 21st issue of Nature Genetics, an international team led by Paz Polak and Jaegil Kim from the Mass General Cancer Center and Broad Institute group of Gad Getz (pictured) found that Signature 3 is also associated with epigenetic silencing of BRCA1 and RAD51C, and with mutations in PALB2 but not in other DNA repair genes. Epigenetic silencing was the most common alteration in African-American patients with Signature 3 whereas protein altering BRCA1/2 mutations were most common among whites. These findings help to explain a puzzling observation in breast cancer genetics and could inform future treatment decisions. 

Recurrent and functional regulatory mutations in breast cancer.

Previous genomic analyses of many tumors have discovered hundreds of cancer genes with mutations in protein-coding regions. By contrast, we do not yet know much about cancer-causing mutations in non-coding regions that regulate important genes. In this June 28 article in Nature, a team of researchers from the MGH Center for Cancer Research and the Broad Institute including first and second authors Esther Rheinbay and Prasanna Parasuraman, co-investigator Andre Bernards and senior author Gad Getz (pictured) describe deep sequencing of 360 primary breast cancers and computational methods to identify significantly mutated promoters. This highly collaborative effort involving several MGH and Broad Institute laboratories found clear signals in the promoters of three genes. FOXA1, a known driver of hormone-receptor positive breast cancer, harbors a mutational hotspot in its promoter leading to overexpression through increased E2F binding. RMRP and NEAT1, two non-coding RNA genes, carry mutations that affect protein binding to their promoters and alter expression levels. This study shows that promoter regions harbor recurrent mutations in cancer with functional consequences and that these hotspot mutations occur at similar frequencies as hotspots in coding regions. The authors also find that we have not exhausted these regulatory mutations yet and are likely to find more by analyzing additional patient tumors in the future. These results emphasize that precision cancer therapy should take account of all types of mutations that affect gene function, be they coding or non-coding. Click here for more information.

Lab picture from left to right: Evmorphia Konstantakou, Danos Christodoulou, Katia Dinkelborg and Othon Iliopoulos. Insert: Arimichi Okazaki, the lead author of the paper.

Glutaminase and poly(ADP-ribose) polymerase inhibitors suppress pyrimidine synthesis and VHL-deficient renal cancers.

In this March 2017 paper in the Journal of Clinical Investigation a group of authors led by Arimichi Okazaki from the lab of Othon Iliopoulos followed up on their previous observations that elevated HIF expression upon loss of the VHL tumor suppressor makes cancer cells dependent on glutamine for synthesis of nucleotides, fatty acids and the detoxification of reactive oxygen species (Metallo CM et al., Nature 2011; 481:380-4 and Gameiro PA et al., Cell Metabolism 2013; 17:372-385). These findings are the basis of an ongoing Phase1/2 clinical trial of the glutaminase 1 (GLS1) inhibitor CB-839 in patients with renal and triple negative breast cancer. In the current paper the authors used metabolic flux analysis and classic tumor biology assays to uncover the mechanism responsible for CB-839-mediated growth suppression of VHL deficient renal cancer cells. Both methodologies highlighted the importance of glutamine for pyrimidine synthesis and DNA replication. Importantly, the fact that cells subjected to GLS1 inhibition showed pronounced DNA replication stress and increased DNA damage suggested that therapeutic synergism might be achieved by combining GLS1 inhibition with inhibition of DNA repair. Indeed, combined inhibition of GLS1 and the DNA repair enzyme PARP caused enhanced DNA replication stress and growth arrest of VHL deficient renal cancer cells. The authors conclude that patients with renal and other HIF-expressing cancers might benefit from treatment with this novel combination of GLS1 and PARP inhibitors and that targeting cancer cell metabolism may lead to novel approaches to treat specific cancers.

Iliopoulos Lab

Decoupling genetics, lineages, and microenvironment in IDH-mutant gliomas by single-cell RNA-seq.

In this March 31 2017 article in Science, a group of investigators from Massachusetts General Hospital and the Broad Institute reports the largest effort to-date in charting brain tumors with single-cell genomic technologies. Analysis of two brain tumor subtypes (IDH-mutant oligodendroglioma and IDH-mutant astrocytoma) has revealed unexpected similarities and differences between these two entities. The authors first compared existing datasets from The Cancer Genome Atlas analyzing 165 bulk RNA-sequencing samples of IDH-mutant oligodendroglioma and IDH-mutant astrocytoma and identified hundreds of genes that are differentially expressed between these two tumor types. As bulk RNA data combines together influences of cancer cell genetics, cancer cell lineages and the tumor micro-environment (TME), the authors generated 14,226 single-cell RNA-seq profiles from 16 patient samples and were able to dissect these differences more precisely. They find that these two entities share similar developmental hierarchies of cancer cells and are both driven by subpopulations of cancer cells with neural stem cell-like programs; they observe that for both tumors, cancer cell differentiation is negatively-correlated with proliferation. While these tumors share similar developmental programs, they differ by genetics and by the composition of the TME. These studies suggest that differentiation therapies could halt tumor growth in IDH-mutant gliomas. The work was led by the team of Mario Suvà (depicted) at MGH and Aviv Regev at the Broad Institute and at MIT. First authors are Andrew Venteicher and Itay Tirosh (depicted). Christine Hebert (depicted), a technician is Suvà’s lab provided tremendous contribution to the work. Click HERE for more information.
Suva Lab

ATR inhibition disrupts rewired homologous recombination and fork protection pathways in PARP inhibitor resistant BRCA-deficient cancer cells.

PARP inhibition selectively kills BRCA1/2-deficient cells, but the efficacy of this therapeutic approach is limited by the onset of drug resistance. In this February 2017 paper in Genes & Development, a group of authors led by Stephanie Yazinski from Lee Zou’s lab (pictured) analyzed how BRCA1 deficient cells develop resistance to PARP inhibition. The paper shows that PARP inhibitor resistance involves the ATR-dependent bypassing of two BRCA1 functions, namely Rad51 loading required for homologous recombination and replication fork protection after fork stalling. Thus, ATR inhibition may overcome PARP inhibitor resistance and offer a therapeutic approach in BRCA-deficient breast and ovarian cancers. Click HERE for more information.
Zou Lab

Functions of Replication Protein A as a Sensor of R Loops and a Regulator of RNaseH1.

In this February 2017 paper in Molecular Cell, co-first authors Hai Dang Nguyen and Tribhuwan Yadav from Lee Zou’s lab (pictured) report a novel function for Replication Protein A (RPA) as a sensor of R loops, a RNA:DNA hybrid transcription intermediate that is a major source of genomic instability. Moreover, RPA interacts and stimulates the activity of RNaseH1, which plays an important role in R loop suppression. Thus, in addition to its well known roles in sensing DNA damage and replication stress, this paper extends the functions of the versatile RPA protein to the suppression of genome instability. Click HERE for more information.
Zou Lab

In this January 2017 paper in Cancer Cell, a group of researchers led by first author Vinod Saladi from Leif Ellisen’s lab (pictured) show that the SWI/SNF chromatin remodeling subunit gene ACTL6A is frequently amplified and highly expressed together with TP63 in head and neck squamous cell carcinoma (HNSCC). ACTL6A and p63 interact and coordinately regulate key genes to control oncogenic YAP activity and patient outcomes in HNSCC. Taken together, the findings define ACLT6A as a potent oncogene and mediator of aggressive tumor behavior in HNSCC. Click HERE for more information.
Ellisen Lab

2019 Research Highlights
2018 Research Highlights
2017 Research Highlights
2016 Research Highlights

Publications By Our Researchers: