Blog

The Early Detection of Cancer Conference- EDx21, hosted by:   Cancer Research UK, OHSU Knight Cancer Institute & Canary Center at Stanford. 

Full daily recap from: Joe Rojas-Burke, Science Writer, OHSU Knight Cancer Institute

 Day 1

Henry Scowcroft, a longtime science writer with Cancer Research UK, gained a shattering new perspective on the disease when his partner Zarah was diagnosed with stage IV bladder cancer at age 36. Scowcroft’s recently published memoir of the caregiving experience is interspersed with details of his science-minded effort to understand the disease that took his partner’s life. In the keynote talk, Scowcroft said he was driven to create something that would help others facing cancer, particularly those not privileged with medical knowledge and expert contacts. Caring for Zarah sensitized him to the burden of tests and treatments on cancer patients. He exhorted researchers to always include a focus on making a difference to the lived experience of people with cancer.
When and where detection matters

Cells with cancer-associated mutations become increasingly common with advancing age, but few ever give rise to invasive tumors. Why not? Phil Jones, Ph.D., at the Wellcome Sanger Institute, showed how emerging tumors can be driven to extinction by competing clonal populations of cells. In studies using mouse esophageal epithelium as a model, it was this competition that eliminated incipient tumors, not immune defenses. With deeper understanding, Jones said it could be possible to manipulate this competition of clones to stop cancer.

Irene Ghobrial, M.D., from Dana-Farber Cancer Institute, also focused on the precursor states, asking whether well-timed interventions can stop the advance to multiple myeloma, an incurable blood cell cancer. Ghobrial is principal investigator of the PROMISE Study, the first to screen healthy people at risk for precursor conditions of multiple myeloma, including monoclonal gammopathy of undetermined significance and smoldering multiple myeloma. The study aims to uncover why some patients progress to myeloma and others do not – and inform the development of screening and preventive treatment strategies. In earlier work, her group identified genetic alterations in smoldering multiple myeloma that could distinguish patients at high risk of progression to multiple myeloma.

Breast cancers that emerge in young women within a few years after pregnancy are strikingly more dangerous. Pepper Schedin, Ph.D., with the OHSU Knight Cancer Institute, detailed how tumor cells are shaped by the postpartum environment to become more prone to metastasis. At the end of lactation, 80 to 90 percent of milk-secreting cells undergo programmed cell death, in a process called mammary gland involution. The early phase of gland involution resembles an acute inflammatory response that subsides and is followed by an immune response similar to what happens during wound healing. Schedin said mammary gland involution is an informative model for studying how factors outside of tumor cells drive cancer development.

In the first lightning talk of the conference, Daniel Muñoz-Espin, Ph.D., from the University of Cambridge, explained how targeting senescent cells could be a way to prevent non-small cell lung cancer progression in people presenting with multifocal primary lesions in their lungs. Such lesions pose a treatment conundrum: most will progress to invasive cancer and there is no good way to stop them. Muñoz said accumulation of senescent cells is a common feature. In a mouse model, drug ablation of senescent cells decreased lung tumor burden and increased survival. In some of the mice, the treatment prevented cancer initiation.
Great debate: Blood is the only detection medium that matters

Minetta Liu, M.D., from Mayo Clinic, made the case for blood – an unpopular position going into the debate with the audience poll showing more than 90% disagreeing. Liu pointed out that existing modalities only screen for a few types of cancer. And some, like colonoscopy, are significantly invasive. Staying on schedule with all the separate screening tests is cumbersome for practitioners and, more importantly, burdensome for patients. Blood-based screening, in contrast, involves a point-of-care blood draw to screen for multiple cancers at once. Results so far suggest blood-based tests are pretty good at identifying the tissue of tumor origin to guide follow-up. Liu gave a whirlwind review of data from several companies readying multicancer, early-detection blood test for broad use, an indicator of the potential for rapid changes in screening practices.

Arguing the contrary, Mark Emberton, M.D., from University College London, asserted that blood tests will only ever serve as a triage tool for selecting patients for more informative imaging tests. He said imaging more reliably delivers consistent results while blood test findings can vary considerably from day to day. He said imaging captures a range of attributes – location, volume, morphology, and more – that can’t be derived from a blood test. And imaging, he said, seems to land at a point in tumor development when cancers have the clear potential to become dangerous, while blood tests might detect too many indolent, non-threatening cases invisible on imaging scans. In the end, Liu swayed 19% to her side while support for Emberton’s position dropped to 81%.

Day 2

Rebecca Fitzgerald, M.D., received the Don Listwin Award for Outstanding Contribution to Cancer Early Detection. She’s known for the development of the Cytosponge, which patients can swallow instead of undergoing an endoscopy. Fitzgerald and her team published work showing the device can increase by 10-fold the identification of Barrett’s esophagus, a precursor to esophageal cancer, compared to standard of care. Fitzgerald talked about the power of collaboration, taking risks and the rise of early detection as a hot topic for researchers in an interview with Cancer Research UK.

Presentations kicked off with a session on emerging technologies for detecting and interpreting early signals of cancer. Stanford University’s Shan Xiang Wang, Ph.D., highlighted the advantages of giant magnetoresistive  (GMR) sensors to look for cancer mutations in circulating DNA, such as EGFR in lung cancer. GMR is more sensitive than fluorescent PCR assays – and less costly, he said. And GMR can show responses to therapy within two weeks, while CT imaging can take two months or more. Wang’s team is also using GMR to find methylated DNA targets in the blood as a way to detect liver cancer in patients with liver cirrhosis, which produces methylation signatures that are difficult to distinguish from cancer.

Can smartphones and wearables help triage the use of expensive and more invasive tools for cancer early detection? Stephen Friend, M.D., Ph.D., from Oxford University, described studies underway with the nonprofit 4YouandMe. Researchers will collect semi-continuous measurements via smartphone, smart ring and smart wristwatch. They will retrospectively analyze relative changes of multi-modal data streams among those who did and did not demonstrate new tumor growth.

Most ovarian cancers are detected dangerously late. Daniel Heller, Ph.D., at Memorial Sloan Kettering Cancer Center, has developed sensors that can be placed inside the uterine cavity, like an IUD, to detect protein biomarkers of ovarian cancer that take too long to show up in blood. In mouse models of ovarian cancer, the carbon nanotube based optical sensors were stable for at least a month after implantation and reliably detected HE4 protein.

In lightning talk, Hannah Brewer, Ph.D., from Imperial College London, gave an update on the Cancer Loyalty Card Study, which is tracking changes in medication purchases that could signal early signs of ovarian cancer. Her team has recruited 117 women diagnosed with ovarian cancer (cases) and 420 women without ovarian cancer (controls). Comparison of loyalty card data of cases and controls has revealed an increase in purchases of pain and indigestion medications prior to diagnosis, suggesting it might be feasible to use such data to prompt medical screening.
 

Great Debate: Emergent properties of the ecosystem are how we’ll understand cancer; a focus on a single cancer cell’s biology is misguided

Single cells can’t tell the story without the context of what’s going on around them, asserted
Aaron Grossberg, M.D., Ph.D., from the OHSU Knight Cancer Institute. Organisms have evolved to respond to localized threats in ways that change systemic physiology, he said. Given that the probability of a positive finding in cancer screening is every low, even if cancer is present, he said that bio-amplification of the signal may be an essential process for detecting cancer. And knowing the prevalence of cancer-associated mutations in non-cancerous tissues, the state of single cells may be misleading without also knowing the state of the microenvironment.

Peter Kuhn, Ph.D., from the University of Southern California, countered that the single cell holds the necessary information, you just need to find the right cell, the needle in the haystack. It is not necessarily the cell that is a cancer cell, it could be a non-cancer cell that is indicative of the cancer, he added. Kuhn changed a few minds; before the debate 75% agreed with Grossberg and 25% disagreed. After, 72% agreed and 28% disagreed.

Early detection in the real world

Starting the day’s final session, Stacey Fedewa, Ph.D., from the American Cancer Society, contrasted cancer screening among women in the United Kingdom, which provides healthcare to all permanent residents, and the United States, where many go uninsured and cancer screening is opportunistic and not organized with coordinated outreach. But in both countries, only about 4 in 10 women were up to date on cervical, breast and colorectal screenings, and 1 in 10 women in both countries said they received no cancer screening.

Livia Giordano, M.D., Ph.D., from Centro di Riferimento per l”Epidemiologia e la Prevenzione Oncologica, made clear the challenges of communication in cancer screening efforts. It is essential to maximize informed choice to participate but also informed choice not to participate in screening, just as it is essential to minimize disinformed participation in addition to minimizing involuntary non-participation. While it’s important to convey knowledge, inform risk perception to help people make choices that fit their values, Giordano said information is not enough. Communicators must consider emotions and trust.

In the real world, the introduction of new technologies can create new health disparities, pointed out Robert Winn, M.D., of VCU Massey Cancer Center. He traced the history of Richmond, Virginia, to illustrate how generations of societal policies have marginalized populations to create health disparities, from segregationist real estate practices, to the concentration of polluting industries in poor neighborhoods. For too long, he said, medicine has avoided directly addressing the social determinants of health.

The day concluded with a lightning talk on a rapid diagnostic service presented by Spencer Robinson, Ph.D., from York and Scarborough Teaching Hospitals NHS Foundation Trust. The service aims to give a single point of access to diagnosis for all patients who have serious nonspecific symptoms. General practitioners refer patients. A triage clinician determines those who will receive a default workup including a CT scan and a trans-nasal endoscopy. Robinson said 336 referrals have resulted in 27 cancer diagnoses, including, in order of prevalence, metastases of unknown origin, pancreatic tumors, gynecologic cancers and lymphomas.

Day 3

Our final day started with a session on models and systems to inform detection. Harvard University’s Benjamin Ebert, M.D., Ph.D., is finding ways to understand the risk of cancer in people with somatic mutations in blood cells, which are extraordinarily common. Access to hundreds of thousands of exomes from peripheral blood is making it possible to trace the path from clonal hematopoesis of indeterminate potential to life-threatening cancer, and to identify events that signal escalating risk. Abnormal blood count is one such event.

Hans Clevers, M.D., Ph.D., at the Hubrecht Institute, detailed the first patient-derived organoid model for cervical cancer, developed in his lab by Kadi Lohmussaar, Ph.D. Ordinary Pap sampling provides enough starting material to grow either healthy organoids or cervical cancer tumoroids. The organoids reproduce gene expression patterns of the organ, as the tumoroids mimic the mutations, gene expression and histology of patient tumors.

Cancer associated fibroblasts, or CAFs, play a role in malignant initiation and progression. Thea Tlsty, Ph.D., from the University of California San Francisco, detailed efforts using bioengineered culture systems to reveal CAF signaling pathways that could be targeted for therapy or early detection. In an air-liquid interface culture system, bronchial epithelial cells grown alongside CAFs acquire squamous characteristics. Adding stress signaling triggers high-grade neoplasia. The next step will be to reproduce the development of full-on cancer.

The session closed with a lighting talk by Carolyn Schutt Ibsen, Ph.D., from the OHSU Knight Cancer Institute. Her team has developed a way to simulate, in an in vitro 3D model, the localized oncogene expression found in early-stage cancers. The system uses lipid-monolayer gas-core microbubbles containing plasmid DNA. Directed ultrasound releases the plasmid DNA from the microbubbles so that only cells in ultrasound focal zone are transfected.

What do we do once we detect early?

Existing cancer therapies have largely been developed to treat advanced tumors. This panel of experts discussed the new therapeutic strategies that will be needed for people diagnosed with very early cancers or pre-cancerous lesions. Chair Charles Swanton, Ph.D., from the Francis Crick Institute/UCL, highlighted a daunting challenge: advances in early detection are revealing that some early-stage cancers can be as deadly as later stage disease. Tom Beer, M.D., at the OHSU Knight Cancer Institute, noted encouraging data suggesting that multi-cancer early detection blood tests might be better at detecting aggressive disease than indolent. Susan Galbraith, M.D., Ph.D., from AstraZeneca, highlighted the need for well tolerated therapies for patients diagnosed with early-stage disease. Avi Spira, M.D., with Johnson & Johnson said localized delivery of drugs could be one way to minimize toxicity. Usha Menon, M.D., University College London, expressed hope for linking screening trials with treatment trials; subjects found to have early disease could be randomized to trials of various treatments. All agreed that understanding the biology of early stage and pre-malignant disease will be critical to finding therapeutic targets for treating early cancers and intercepting pre-malignant disease with less toxic treatments.

Great debate:   It’s time to give up on human hypothesis-driven research; massive, multimodal data mined by AI is the only way we’ll truly unravel the complexity of cancer early detection

Relying on hypothesis-driven research is like wandering through an infinite jungle hoping to stumble on the correct path, asserted Bissan Al-Lazikani, Ph.D., Institute of Cancer Research/MD Anderson Cancer Center. Early detection faces a signal-to-noise ratio problem, she said, that can only be solved by machine. She pointed to the development of BRAF inhibitors. BRAF emerged as a worthwhile target from computational analysis of massive amounts of DNA sequencing data, not from a scientist’s hypothesis about that particular signaling pathway.

Xin Lu, FRS F.Med.Sci., with Ludwig Cancer Research and the Oxford Centre for Early Cancer Detection, countered that AI-generated results lack meaning if they can’t be explained by a scientific hypothesis. A machine learning algorithm may find complex patterns linked to early cancer, but without a hypothesis to explain them, it’s impossible to validate the truth, Lu said. And AI requires a starting point and training data to learn from, which Lu said requires a human-generated hypothesis about the way the world works.

Going in, 82% of the audience sided with Lu, and 18% with Al-Lazikani. After the debate, Lu’s support dropped to 54% and 46% agreed with Al-Lazikani.

It’s been a productive three days!

On behalf of Cancer Research UK, the OHSU Knight Cancer Institute and the Canary Center at Stanford, thank you for joining us for the 2021 Early Detection of Cancer Conference.

Farewell,

The Early Detection of Cancer Conference events team

Recapping day one

We’re glad you made it to the 2020 Early Detection of Cancer Conference. We’re looking forward to more eye-opening presentations, well-argued debates and networking opportunities. Before jumping back into it, here’s a quick recap of day one:

Caroline Dive (University of Manchester, CRUK Manchester Institute) brought us up to speed on efforts to improve lung cancer screening by combining CT imaging with a liquid biopsy. (Her team is running an observational cohort study to test whether blood biomarkers can detect lung cancer recurrence earlier than standard of care clinical surveillance.) The field has no shortage of potential biomarkers to choose from, including: circulating tumor cells, tumor DNA, RNA, and tumor educated platelets. Progress, Dive said, will hinge on a deeper understanding of early disease biology and pre-clinical models that more accurately represent the early stages of cancer.

To that end, Anton Berns (Netherlands Cancer Institute) highlighted the promise of autochthonous tumor models, that is, tumors induced in lab animals, in which it is possible to study early tumor formation in the presence of an intact immune system. With such models, researchers can switch particular oncogenes and tumor-suppressor genes on or off in a given tissue and compare cancer development. Berns said his team’s mouse models closely recapitulate the phenotype of human cancers including small cell lung cancer, and may help identify specific early biomarkers of dangerous tumors.

The first session closed with two lightning talks: Naoki Oshimori (OHSU Knight Cancer Institute) described how a mouse model of squamous cell carcinoma enabled his team’s discovery of a signaling loop between tumor-initiating cells and nearby non-cancer cells that generates the niche microenvironment that is required for invasive progression and drug resistance. Jennifer Munkley (Newcastle University Biosciences Institute) gave an update on the GlycoScore blood test for prostate cancer, which looks for specific glycans (sugars that attach to proteins, lipids, and other glycans on cells). Tested in more than 600 patient samples, a three-glycan test distinguished between benign tissue and prostate cancer with high sensitivity and specificity, she said.

Reflecting on COVID-19 The COVID-19 pandemic, as in all of medicine, has posed severe challenges for cancer screening. Participants in a special panel discussion called out opportunities the pandemic has created. When it became unfeasible for patients to visit the clinic for melanoma screening, Sancy Leachman (OHSU Knight Cancer Institute) and colleagues came up with an alternative: dermatoscopes that attach to a mobile phone, which high-risk patients can borrow and transmit images of suspicious lesions. It’s become a permanent option for rural patients and those who can’t easily travel. For patients with throat symptoms calling for endoscopy, Rebecca Fitzgerald (Cambridge University, MRC Cancer Unit) said her center began cautiously testing an alternative: the Cytosponge, a small mesh sponge within a soluble gelatin capsule that is swallowed and retrieved to collect esophageal cells. Kevin Monahan (St. Marks Hospital) said his team learned the cost of halting colonoscopy procedures and is working to safely maintain the service for symptomatic patients even if a pandemic second wave hits hard. Jackie Shannon (OHSU Knight Cancer Institute) said the pandemic has brought wide attention to long entrenched inequalities and health disparities, perhaps enough to drive much-needed policy changes and enduring efforts to reach underserved populations.

Leveraging risk stratification Cancer screening intensity should be matched to an individual’s risk of getting cancer. Jon Emery (University of Melbourne) described efforts to use genetic testing to help patients make informed decisions on colorectal cancer screening. He said it’s looking feasible to start to implement genetic risk stratification in the general practice setting. In the future, results will be even better with decision support tools that include risk factors such as diet, smoking, screening history, and medication use.

Julia Hipsley-Cox (University of Oxford) and colleagues are drawing upon the UK health system’s deep and detailed patient records to develop risk stratification algorithms to target cancer screening resources to people at highest risk and most likely to benefit from interventions (you can check them out at www.qcancer.org). Other tools are designed to be integrated into electronic medical record systems.

The day ended with two more lightning talks: Rebecca Landy (National Cancer Institute) noted a huge disparity in lung cancer screening guidelines: 32% of African Americans who developed lung cancer would have been eligible for CT screening, compared with 56% of whites. She showed how an individualized risk calculator (the LYFS-CT model) can effectively eliminate this disparity. Tom Callender (University College London) presented findings on the impact of MRI prior to biopsy on age-based and risk-tailored screening for prostate cancer.

Thank you for joining us. Please accept our sincere apologies for the technical challenges. Don’t forget: the video library will be updated each day with recordings of the meeting sessions.

Recapping day two

We have three more sessions lined up, but first here’s a recap of Wednesday.

Risk-tailored screening is a way to fit the intensity of testing to an individual’s risk of getting cancer. Hilary Robbins (International Agency for Research on Cancer) focused on the job of generating evidence that will be needed to establish risk-tailored cancer screening, presenting examples from lung cancer and breast cancer. Randomized clinical trials are not the way forward, she said, given the large numbers of subjects and lengthy follow-up needed just to answer a limited number of questions in only one context.

A cancer blood test developed by GRAIL, Inc., is being evaluated for its ability to detect more than 20 types of cancer and predict tissue of tumor origin. GRAIL Vice President Eric Fung highlighted the clinical studies that have led the company to focus on DNA methylation patterns for its multi-cancer early detection test undergoing a multicenter clinical trial due for completion in early 2021.

Two lightning talks closed the session: Amelie Lutz (Stanford University) is developing an ultrasound guided molecular imaging method for detecting ovarian cancer using microbubbles that target tumor angiogenesis. Stefano Avanzini (Stanford University) is using mathematical models to estimate the size tumors must reach to become detectable by tumor DNA circulating in blood. (For lung cancer, he estimates a median tumor detection size of 2 cm, which is a 43% decrease compared with the median size of diagnosed cancers in the SEER database.)

KEYNOTE TALK

Dinah S. Singer (National Cancer Institute) began with a rundown of the NCI’s response to COVID-19 pandemic, from virus-focused research initiatives to the ways the agency is flexing to support grantees. She concluded with an overview of the cancer early detection programs the agency has underway, such as the Early Detection Research Network (now focusing on AI and machine learning to integrate omic data to find biomarkers), and The Human Tumor Atlas Network (HTAN), a massive effort to map the complex ecosystems of cancer – and pave the way for advances in prevention, early detection and treatment.

GAMBHIR HONORED WITH DON LISTWIN AWARD

Sanjiv Sam Gambhir was an internationally recognized pioneer in molecular imaging who dedicated his career to developing methods of early disease detection. The director of the Canary Center at Stanford died of cancer on July 18. He was honored with the Don Listwin Award in a ceremony with heartfelt and moving remembrances from Utkan Demirci (Stanford University) and Iain Foulkes (Cancer Research UK). The Listwin Award was established last year to recognize a sustained contribution to, or singular achievement in, the cancer early detection field. The award is named in honor of Don Listwin, founder and chairman of the Canary Foundation.

Recap of the final day

The third and final day kicked off with some eye-opening updates from the world of AI and machine learning.

Lily Peng and Sunny Jansen (Google Health) expounded on three overlooked requirements for building successful AI models: data of high quality, not just quantity; human-centered usability, not just model accuracy; cost-effectiveness, not just excellent performance.

AI systems are becoming adept at reading radiology images and pathology slides to correctly classify lesions as cancer or benign. Parag Mallick (Stanford University) explained how tools such as saliency mapping are making it possible to understand how the machines reach their conclusions – building confidence and potentially revealing biological insights. He also showed examples of AI tools for biomarker discovery that extract and create knowledge from massive, unstructured data sets.

Two lightning talks concluded the session: Freya Woods (Swansea University) showed how AI can improve the sensitivity and specificity of cancer detection by Raman spectroscopy, which her group is developing as a triage tool in the diagnosis of colorectal cancer. Rawen Kader (University College London) and colleagues have developed a neural network to assist real time decision-making during colonoscopy by classifying polyps as pre-cancerous or not, with a randomized clinical trial in the offing.

GREAT DEBATES

Should genomic risk stratification be part of early detection? Gareth Evans (Manchester University) made the case that it must, noting that polygenic risk scores robustly predict risk for several common cancers and can be used to fit the intensity of screening to a person’s risk of getting cancer. Cristian Tomasetti (Johns Hopkins University) argued that, while genomic risk stratification is useful for some cancer types, many others have no known inherited factors. He asserted that the development of affordable and minimally invasive multi- cancer blood tests will reduce the need for genetic risk stratification. Before the debate, 60% of meeting attendees agreed with Evans, and 40% agreed with Tomasetti. The ratio shifted to 50:50 after.

Before approving new early detection approaches for clinical use, should we require evidence of a cancer-specific mortality benefit from at least two randomized controlled trials? Harry De Koning (Erasmus University Medical Centre) pointed to the conflicting findings of clinical trials of screening methods such as PSA for prostate cancer to make the affirmative case. Steve Skates (Harvard University) asserted that requiring such evidence unnecessarily delays the use of early detection advances, and costs too much, when there are faster and less costly trial endpoints, such as reduction in late-stage diagnoses. In the poll of meeting attendees, agreement with De Koning dropped from 32% pre- debate to 20% percent after, with many deciding that it’s too much to ask for randomized trials showing mortality benefit.

On behalf of the Canary Center at Stanford, Cancer Research UK and the OHSU Knight Cancer Institute, thank you for joining us for the 2020 Early Detection of Cancer Virtual Conference. With luck, we’ll be able to meet in person at next year’s meeting. For now, the organizing committee has decided to be optimistic and start preparing for an in-person gathering in London.