There is no debating that technologies and approaches centered on prevention and detection are key to effectively treating cancer and improving survival rates.
For prevention’s part – which can be defined simply as “actions taken to lower the risk of getting cancer” – the “proof is in the pudding.” According to the World Health Organization, between 30 and 50% of cancers can be prevented by avoiding risk factors (avoiding tobacco use, maintaining a healthy weight, regular exercise, etc.) and implementing existing evidence-based prevention strategies.
Similarly, early detection has been shown to dramatically improve a patient’s prognosis and survival rate following a cancer diagnosis. For example – per the American Cancer Society, the 5-year relative survival rate for localized breast cancer (cancer that has not spread outside the breast) is a near-perfect 99 percent.
Clearly, prevention and detection matter and the cancer community has made tremendous strides in recent years to bring prevention education and tools and enhanced detection methods to more and more people. However, access remains the primary challenge. This is especially true of historically underserved communities.
But with more tools and technology, we can continue to close this gap, which begs the question, what’s new in cancer prevention and detection? It turns out, quite a lot. And that’s great news for everyone.
This week, we’ll examine a few recent, key innovations and initiatives that are continuing to move the needle on early detection. In the next installment, we’ll share some of the latest on the prevention front.
Liquid Biopsy Can Enhance Diagnosis and Guide Personalized Care of B-cell Lymphoma
The most common cancer of the lymphatic system, B-cell lymphoma accounts for 85% of all non-Hodgkin lymphoma. The American Cancer Society estimates 80,600 people will receive a non-Hodgkin lymphoma diagnosis in 2024.
The rate of recurrence for patients with aggressive B-cell lymphoma is approximately 30%.
One of the primary challenges related to the diagnosis of B-cell lymphoma is that the current methods for detection are not reliably accurate. A problem that researchers at the University of Helsinki and Helsinki University Hospital recently set out to solve.
Historically, risk profiling for B-cell lymphoma was performed based on factors such as age, overall health, and disease stage. Relying on these factors often meant that those at highest risk for the disease went undetected. In addition, inadequate tissue samples could result in inaccurate diagnoses.
To address these challenges, researchers examined how blood samples from lymphoma patients could be used to improve both the diagnosis and treatment of B-cell lymphoma. They studied blood samples from 109 patients with aggressive B-cell lymphoma, collected before treatment, mid-treatment and at the end of treatment. Specifically, the team analyzed the levels of 1,400 proteins, along with the patients’ clinical data for comparison purposes.
Findings
Through the study, the team discovered an inflammatory protein profile they say is associated with low rates of survival, inflamed tumor tissue, and “tumor burden,” which refers to the total number of mutations found in the DNA of cancer cells.
Researchers also found that they could classify different subtypes of B-cell lymphoma based on the protein profiles obtained from blood samples. They underscored the fact that the monitoring of protein data can in turn enable the monitoring of patients’ response to treatment – an important observation with implications for targeted treatment and care.
“We found that the protein profiles of blood samples can help direct care to the patients who will most benefit from it. This technique would significantly boost personalized care, as it takes into account both the characteristics of tumor tissue and the patient’s response to the disease,” said Professor Sirpa Leppä from the University of Helsinki and HUS Helsinki University Hospital in a news release.
The team also concluded that protein profiles could be used to increase the accuracy of diagnoses “in cases where a tissue specimen alone is not enough,” and can also be helpful in follow up care and monitoring for recurrences.
“In practice, this could mean that it would be possible to monitor any potential relapses with the help of blood samples instead of imaging,” said Doctoral Researcher Maare Arffman from the University of Helsinki.
Next Steps
The research team now plans to further test the feasibility of protein profiles in cancer care through clinical trials; specifically utilizing blood samples in profiling lymphoma patients in order to inform personalized plans of care.
“The proteins and circulating tumor DNA in blood samples have enormous potential for improving cancer diagnostics,” said Leppä. “Refining patient risk profiling, and guiding treatment decisions.”
Researchers Use AI to Discover New Cancer Imaging Biomarkers, Improve Detection
The latest and greatest technology continues to drive innovation in the cancer sphere. This is perhaps nowhere more evident than in a recent study published in Nature Machine Intelligence, which details how Mass General Brigham researchers are using – you guessed it – AI to uncover new cancer imaging biomarkers that could, “transform how patterns are identified from radiological images;” an approach that would dramatically impact both the early detection and treatment of cancer.
The technology behind it all, foundational models, is the same tool that the popular ChatGPT chatbot software is built upon. But whereas ChatGPT uses language models, the researchers at Mass General have developed image models.
Specifically, they built their foundational model using a dataset of 11,467 images of abnormal radiologic scans. These images allowed the model to identify patterns that predict anatomical site, malignancy, and prognosis “across three different use cases in four cohorts.”
The model proved especially effective in more complex cases where only limited data may be available.
“Our findings demonstrate the efficacy of foundation models in medicine when only limited data might be available for training deep learning networks, especially when applied to identifying reliable imaging biomarkers for cancer-associated use cases,” said senior author Hugo Aerts, PhD, director of the Artificial Intelligence in Medicine (AIM) Program at Mass General Brigham.
In their findings, researchers stated that the model, “provided superior performance for anatomical lesion site classification on average and across individual anatomical sites, even when very few training samples were available for that site.”
The model also outperformed all other baseline approaches in respect to malignancy prediction.
The team at Mass General concluded that the model serves as a powerful and accurate tool for detecting cancer imaging biomarkers, especially in cases where limited information is available.
The researchers are also optimistic that this approach can uncover even more biomarkers, spurring further innovation in the detection and treatment of cancer.
“Given that image biomarker studies are tailored to answer increasingly specific research questions, we believe that our work will enable more accurate and efficient investigations,” said first author Suraj Pai of the AIM Program.
In a true spirit of innovation and open-source frameworks, the team at Mass General Brigham shares their model and reproducible workflows; stating it is their hope that “more studies can investigate our methods, determine their generalizability and incorporate them into their research studies.”
Cancer Moonshot Announces Commitments from Employers and Labor Unions to Make Cancer Screenings More Accessible for American Workers
Over the past year, we’ve covered the Cancer Moonshot’s initiatives extensively, as well as the program’s recent recognition of Breaking Cancer News and Teen Cancer America.
Fortunately, President Biden and the Moonshot team show no signs of slowing following an announcement that some of the nation’s largest and most influential companies will be implementing various cancer screening programs and initiatives for employees.
The list of companies taking steps to provide accessible cancer screening to employees includes some heavy hitters – Target, United Airlines, Lyft, Genentech, Amazon, Gilead, Hasbro, and more. The programs being put in place by these companies range from Genetech’s women’s health screening day to Target’s “incentives to earn up to $500 annually by receiving recommended preventive care.”
The announcement follows an April meeting in which the Moonshot convened a diverse group of employers to “advance knowledge-sharing and best practices to make cancer screenings more accessible for their employees.”
