Collaboration: It’s a hot topic in cancer research these days, particularly since Vice President Joe Biden took on the Cancer Moonshot initiative. Scientists and institutions across the country are joining forces to accelerate cancer research like never before, and Fred Hutchinson Cancer Research Center is deep in the mix.
What does this mean for breast cancer, a disease that strikes 250,000 women and men — and kills nearly 41,000 of them — each year? We checked in with breast cancer researchers at Fred Hutch and Seattle Cancer Care Alliance to see what they’re particularly enthusiastic about as Breast Cancer Awareness Month draws to a close.
Collaborations and combinations
Along with research institutions pooling ideas and talents, scientists are also exploring the advantages of combining different drugs and treatments.
“Combinations are the key,” said Fred Hutch clinical researcher and Seattle Cancer Care Alliance oncologist Dr. V.K. Gadi, at the recent Northwest Metastatic Breast Cancer Conference.
Of particular interest are “biologics,” engineered drugs like the breast cancer therapeutic Herceptin that target certain features found within a tumor.
“Biologics are not things that grow on a tree. We didn’t find them in the soil,” said Gadi. “These things were engineered from the ground up. We found a problem in a cancer cell that was unique to that cancer cell, and we went off and designed a therapy for it. There are a ton of these medicines and we’re getting to the point where we’re not just using them one by one by one.”
Researchers are currently figuring out how to combine these biologics in new — and less toxic — ways. Combination therapies make sense, particularly in breast cancer, because tumors often have a host of genetic mutations. The combined agents can attack and overwhelm the cancer before it can mutate yet again.
“Really, these are where all of our hopes lie,” said Gadi. “We’re learning how we can safely combine these things so we don’t have overlapping toxicities.”
Several combination therapy trials are currently being conducted at Fred Hutch.
“Our ATEMPT trial hopes to provide a less toxic and perhaps more effective therapy to very early stage patients,” Gadi said. The plan is to substitute a new biologic therapy for chemotherapy in patients with stage 1, HER2-positive breast cancers — early cancers that have a certain overexpressed protein.
The randomized Phase 2 trial will compare the safety and effectiveness of two drug regimens. Half the patients will receive Kadcyla, a new biologic known as an antibody drug conjugate, or ADC, that combines Herceptin with a very toxic chemotherapy known as emtansine. The other half will receive the chemotherapy agent Taxol in combination with Herceptin, a protocol that is considered standard of care for early HER2-positive breast cancer.
Gadi said the new biologic will be targeted directly to the cancer cell, while “generally” sparing normal tissues. Kadcyla is currently approved by the U.S. Food and Drug Administration for metastatic HER2-positive breast cancers only.
Two other promising trials, one for early-stage and one for late-stage breast cancer, also combine two or more therapies. They are being conducted by SWOG, a long-standing worldwide network of physician researchers who design and conduct cancer clinical trials. Fred Hutch houses its Statistical Center.
The early-stage breast cancer trial SWOG 1207 is a randomized Phase 3 trial that will determine if the targeted therapy everolimus — or Afinitor — can improve disease-free survival in patients with aggressive ER-positive breast cancer, Gadi said. This trial will combine Afinitor with various forms of endocrine therapy, which blocks the production of estrogen, the fuel for so-called hormone-positive breast cancers.
The second trial, SWOG 1416, is designed for metastatic breast cancer patients with triple- negative breast cancer, i.e., cancer that is negative for hormone receptors and the HER2 receptor. “This is a huge and important national study that is a direct offshoot of a very successful trial we completed entirely here,” Gadi said. That study, published last June, found “encouraging response rates” with a combination of veliparib (a type of targeted therapy known as a PARP inhibitor) and the chemotherapy agents cisplatin and vinorelbine. The new Phase 2-3 trial will further explore this response in metastatic triple-negative breast cancer patients as well as those with BRCA mutations.
Fred Hutch breast cancer researchers are also participating in several immunotherapy trials, Gadi said, including trials for immune system boosters called checkpoint inhibitors and breast cancer vaccines. Researchers involved with these trials include Drs. Jennifer Specht, Lupe Salazar, Sasha Stanton, Stan Riddell, Laura Chow and Peggy Porter.
“We have several trials in process that combine immunotherapies from other diseases for use in triple-negative breast cancer,” he said. “We have a first-time human CAR T-cell trial for triple-negative breast cancer. Dr. Lupe Salazar has a trial using a ‘super’ trastuzumab [Herceptin] called margetuximab that is capable of higher level engagement of the immune system. And we also have trials for patients with HER2-positive disease that really juice the immune system to improve responses. The SOPHIA trial is open now and the KATE2 trial will be open soon.”
The KATE2 trial, which will be activated in late November or early December, is a trial of the biologic Kadcyla combined with the PD-L1 antibody checkpoint drug, atezolizumab.
“We’ve got a lot of angles covered and most of these trials are only available at our site in the Pacific Northwest,” Gadi said.
(More information about Fred Hutch clinical trials and Seattle Cancer Care Alliance trials.)
Collaborating with coders and cloud-computing giants
Mammograms are the best way to find breast cancer early, when it’s easier to treat. Unfortunately, the screening process is far from perfect.
Currently, out of every 1,000 women screened, 100 are called back for additional mammograms, biopsies or other tests; five are diagnosed with breast cancer. Unfortunately, some women go through these callbacks and biopsies needlessly due to false positives. In others, cancers are missed because of human or technical error.
Fred Hutch researcher and SCCA radiologist Dr. Christoph Lee is hoping these problems will become a thing of the past, thanks to a recently launched Digital Mammography DREAM Challenge.
Co-created by Lee, the Challenge pits teams of researchers and coders against each other in a “collaborative competition” to fast-forward breast screening technology by using machine learning. The latest buzzword in advanced computing, machine learning is a form of artificial intelligence where computers are taught to self-adjust their programming without human help. So far, nearly 800 participants from around the world have registered, each vying for cash prizes and the opportunity to help millions of women.
“The Digital Mammography DREAM Challenge has put up $1.2 million in prize money to coders to come up with a machine- or deep-learning solution to improving accuracy in screening mammography,” he said.
DREAM Challenges — the acronym stands for “Dialogue for Reverse Engineering Assessments and Methods” — are community challenges that tackle questions in biology and medicine by tapping the combined brain power of university researchers, nonprofits, and experts in the technology, biotechnology and pharmaceutical industries. You might call it crowd-sourcing for geniuses. These open-science challenges are just the type of silo-busting, uber-collaborative research Vice President Biden has repeatedly lauded since the Cancer Moonshot launched. This Challenge, in fact, was featured by the White House during the June 28 Moonshot Summit.
Organized by Fred Hutch-based Sage Bionetworks, IBM Research, Group Health Research Institute, Seattle Cancer Care Alliance, the National Cancer Institute and others, the Challenge’s results will be placed into the public domain, with the algorithms made available to researchers. Cloud computing — storing and processing data via the internet as opposed to a local server — is being provided by Amazon Web Services Inc. and the IBM Watson Health Cloud.
Lee, who also serves as the Digital Mammography DREAM Challenge clinical adviser, said radiologists are currently unable to see up to 16 percent of breast cancers when reading mammograms. He hopes machine learning will change that.
“We’re trying to extract more useful information out of our imaging,” Lee said. “It’s sort of a ‘brave new world’ type of thinking, combining data analytics and medicine. We’re not looking to replace radiologists but rather looking for ways that machine learning can augment human skills.”
Participating teams will submit models based on 640,000 digital mammograms from more than 86,000 patients — whose names have been scrubbed from the data — as well as clinical and demographic information for those patients, including their race and/or ethnicity, weight, breast density, family history and history of disease or biopsy. They will then use the data to look for patterns, make connections and eventually develop a tool to bump up the accuracy and reliability of digital mammography.
Officially launched in June, Challenge results aren’t expected until next spring, Lee said. Until then, he’s hopeful the friendly coding competition will eventually save lives.
“Detecting invasive breast cancers at earlier stages on screening mammography can lead to decreased morbidity and mortality among women,” he said.
Filling in crucial data gaps
Public health researcher Dr. Ruth Etzioni, a Fred Hutch biostatistician who works in modeling for breast cancer screening, is particularly excited about a grant she just submitted to the National Cancer Institute, or NCI.
Her proposed project? Filling in the information gaps about breast cancer recurrence in the SEER Cancer Registry (Surveillance, Epidemiology and End Results). Etzioni said the NCI has itself identified the absence of recurrence information in the cancer registry as a “critical gap” and a priority for future work.
“Although the SEER Registry provides extensive information about disease at diagnosis and some data on primary treatment and survival, there is no information about recurrence,” she said.
Patients, too, have called for more accurate recurrence data; so far, more than 3,500 have signed a petition launched just a month ago by the advocacy group MBC Alliance (MBC stands for metastatic breast cancer) that calls for SEER to “start counting ALL people living with metastatic breast cancer.”
If funded, Etzioni will use a cohort of the SEER Registry known as the Puget Sound Cancer Surveillance System, or CSS, which collects population-based data on cancer incidence and survival in 13 western Washington counties. Her plan is to use medical claims data and patient-reported outcomes, or PROs, to eventually develop, validate and deploy a scalable, automated approach for identifying cancer recurrence.
Etzioni was inspired to use patient-reported data after hearing about the success of the Metastatic Breast Cancer Project at the Broad Institute. “They have recruited nearly 3,000 patients in just a year via social media and the web,” she said. “I think the direct-to-patient approach holds a lot of promise.”
Etzioni is excited about the prospect of collaborating with other institutions, as well, specifically Swedish Cancer Institute, Group Health, and the University of Kentucky. The Kentucky program houses its own SEER Registry; Etzioni hopes to use it to show that the new approach can scale up across multiple registries.
“There’s a national effort to expand collaborations across sites that house valuable data for cancer research,” she said. “We very much want to build bridges and be a part of that effort.”
Etzioni has named her proposed project ReCAPSE: Recurrence using Claims and PROs for SEER Expansion.
“It’s meant to rhyme with ‘relapse,’” she said. While the initial focus is on breast cancer, the hope is to expand the project to include all SEER registries and all cancers.
“Folding information about disease recurrence into the various cancer registries is critical to improving cancer surveillance, understanding the burden of cancer and studying the effectiveness of cancer treatments” she said. “This will be a key step forward.”
For information about participating in a cancer clinical trial, please contact Fred Hutch’s clinical care partner, Seattle Cancer Care Alliance, or email intake@seattlecca.org. Another good resource about clinical trials is the National Cancer Institute’s Cancer Information Service, which can be reached at 1.800.4.CANCER. The phone line is staffed from 8 a.m. to 8 p.m. ET Monday through Friday.
Diane Mapes is a staff writer at Fred Hutch. She has written extensively about health issues for NBC News, TODAY, CNN, MSN, Seattle Magazine and other publications. A breast cancer survivor, she blogs at doublewhammied.com and tweets @double_whammied. Email her at dmapes@fredhutch.org.
