A breakthrough study published in The American Journal of Pathology describes a new machine-learning model that may improve accuracy in early diagnosis of hepatocellular carcinoma and monitoring the impact of treatment.
Early diagnosis of hepatocellular carcinoma (HCC) – one of the most fatal malignancies – is crucial to improve patient survival. In this breakthrough study, investigators report on the development of a serum fusion-gene machine-learning model. This important screening tool may increase the five-year survival rate of patients with HCC from 20% to 90% because of its improved accuracy in early diagnosis of HCC and monitoring the impact of treatment.
HCC is the most common form of liver cancer and accounts for around 90% of cases. Currently, the most common screening test for the HCC biomarker, serum alpha-foetal protein, is not always accurate, and up to 60% of liver cancers are only diagnosed in advanced stages, resulting in a survival rate of only around 20%.
Lead investigator Jian-Hua Luo, MD, PhD, Department of Pathology, High Throughput Genome Center, and Pittsburgh Liver Research Center, University of Pittsburgh School of Medicine, explained: “Early diagnosis of liver cancer helps save lives. However, most liver cancers occur insidiously and without many symptoms. This makes early diagnosis challenging. What we need is a cost-effective, accurate, and convenient test to screen early-stage liver cancer in human populations. We wanted to explore if a machine-learning approach could be used to increase the accuracy of screening for HCC based on the status of the fusion genes.”
In the search for a more effective and efficient diagnostic tool to predict non-HCC and HCC cases, investigators analysed a panel of nine fusion transcripts in serum samples from 61 patients with HCC and 75 patients with non-HCC conditions using real-time quantitative reverse transcription PCR (RT-PCR). Seven of the nine fusions were frequently detected in HCC patients. The researchers generated machine-learning models based on serum fusion-gene levels to predict HCC in the training cohort, using the leave-one-out cross-validation approach.
A four fusion gene logistic regression model produced an accuracy of 83% to 91% in predicting the occurrence of HCC. When combined with serum alpha-foetal protein, the two-fusion gene plus alpha-foetal protein logistic regression model produced 95% accuracy for all the cohorts. Furthermore, quantification of fusion gene transcripts in the serum samples accurately assessed the impact of the treatment and was able to monitor for the recurrence of the cancer.
Dr. Luo commented, “The fusion gene machine-learning model significantly improves the early detection rate of HCC over the serum alpha-fetal protein alone. It may serve as an important tool in screening for HCC and in monitoring the impact of HCC treatment. This test will find patients who are likely to have HCC.”
Dr. Luo concluded, “Early treatment of liver cancer has a 90% five-year survival rate, while late treatment has only 20%. The alternative to this test is to subject every individual with some risk of liver cancer to imaging analysis every six months, which is very costly and ineffective. In addition, when imaging results are ambiguous, this test will help to differentiate malignant versus benign lesions.”
One of the biggest stories in HIV in the last year was that a class of medicines called statins could help reduce cardiovascular disease in people living with the virus. In response, treatment guidelines in the United States were quickly updated, but the picture is more complicated in South Africa. Spotlight’s Elri Voigt explores why the case for widespread use of statins by people living with HIV is less compelling in South Africa than in some other countries.
People living with HIV, provided they are stable on antiretroviral therapy, are affected by the same diseases as those who don’t have HIV, including cardiovascular disease, says Professor Mpiko Ntsekhe, head of Cardiology at Groote Schuur Hospital in Cape Town.
The key difference, he says, is that although both groups of people get the same spectrum of diseases, people living with HIV get those diseases more frequently and earlier. One way to think about this, he explains, is to imagine twins who are identical in every way except one is living with HIV. The twin living with HIV is more likely to get cardiovascular disease than the other twin.
And these differences can be substantial. Current evidence shows that people living with HIV have a twofold increased risk of developing cardiovascular disease compared to people not living with HIV, says Professor Hans Strijdom. He is the Head of the Division of Medical Physiology and Deputy Director of the Centre for Cardio-Metabolic Research in Africa (CARMA) at Stellenbosch University. The cardiovascular risk attributable to HIV, Strijdom adds, is now believed to be equivalent to that posed by traditional risk factors such as smoking. This prompted an editorial in 2018 in one of the top cardiovascular journals, Circulation, advocating for HIV to be recognised as a major cardiovascular risk factor.
He explains that people living with HIV who are stable on treatment are living longer, making them susceptible to the normal risk posed by older age. They also have “modifiable risk factors, in other words lifestyle risk factors”, like a higher smoking and alcohol use incidence, as well as increasing rates of being overweight and obesity. Strijdom says that living with HIV, even when someone is stable on treatment, causes low-grade inflammation, which over time increases a person’s risk for cardiovascular disease. “That all in combination are the current theories [of] why we think that they have a bigger risk of cardiovascular disease,” he says.
Important study findings
Arguably, the biggest news from last year’s International AIDS Society (IAS) Conference in Australia was findings from a study on heart disease in people living with HIV. The trial, called REPRIEVE, showed that a class of cholesterol-busting drugs called statins can prevent a lot of cardiovascular disease events in people living with HIV whose cardiovascular disease (CVD) risk score meets a certain threshold. Spotlight previously reported on these findings, which showed that compared to placebo, daily treatment with 4mg oral pitavastatin – a specific statin – led to a 35% reduction in major adverse cardiovascular events (MACE) in people living with HIV classified to be at risk of cardiovascular disease.
When the findings were presented at the IAS conference, the study’s principal investigator, Dr Steven Grinspoon, said that while the researchers still have to assess more of the data collected to get a clearer picture of things, like the mechanisms driving cardiovascular disease across regions and conduct additional sub-group analyses, the study has already shown that using pitavastatin can save lives.
These sub-group analyses were discussed in greater detail at the Conference on Retroviruses and Opportunistic Infections (CROI) held in Denver in March this year. For the most part, the use of pitavastatin in the manner prescribed by REPRIEVE was considered a huge success, and the United States has since changed its guidelines to include the use of statins in the primary prevention of atherosclerotic cardiovascular disease.
Why it is different in South Africa
However, for low-and-middle-income countries like South Africa, the case for pitavastatin might not be as clear-cut. In fact, a panel discussion at CROI was dedicated to exploring the implications of the REPRIEVE findings for such countries.
Ntsekhe, who was a speaker on the CROI panel, tells Spotlight that data from REPRIEVE’s sub-group analyses reveal there was a striking difference in event rates – which in the case of the study are MACE in those who were getting the placebo – by country income status. He explains that as predicted in high-income countries, the event rates were high, while in low-and-middle income countries – particularly in Sub-Saharan Africa – event rates were very low.
He says one of the reasons for the difference in event rates was that the screening tool used in REPRIEVE worked well to identify those people living with HIV who might benefit from pitavastatin in high-income countries like the United States, but it did not work well in Sub-Saharan Africa.
This means using pitavastatin as part of a primary prevention strategy is a much more effective intervention in high-income countries than in low-and-middle income countries like in Sub-Saharan Africa because the cardiovascular disease profile is so different.
Ntsekhe explains the term cardiovascular disease itself is broad and all-encompassing and there are many forms, including valve disease, heart muscle disease, and vascular disease. The dominant form of cardiovascular disease in the high-income countries (which he refers to as the Global North) is known as atherosclerotic cardiovascular disease, which is characterised by a build-up of fatty deposits and plaque in the arteries.
In Sub-Saharan Africa though, Ntsekhe says “atherosclerotic cardiovascular disease is but one of many forms of cardiovascular disease”, taking the fourth or fifth place in the ranking of types of major heart disease.
Research conducted in high-income countries don’t always take differences in disease burden into account, according to Ntsekhe. This means that interventions researched in high-income countries and shown to be effective in that context won’t necessarily work as well in low-and-middle income countries like South Africa.
Strijdom concurs that while results from REPRIEVE in the global context were a game-changer, the findings are not easily transferable to South Africa’s context because pitavastatin is mainly aimed at reducing “bad cholesterol” and coronary artery disease (also called atherosclerosis).
‘Taking money away’
During the panel discussion at CROI, Ntsekhe asked whether Sub-Saharan Africa could justify taking money away from other health programmes that work in order to invest in pitavastatin.
“I said basically what should be a priority for us is a) finding tools that can better identify those at risk and b) continuing to focus on what our local data suggests are the priority areas,” Ntsekhe says.
“If your entire prevention strategy is aimed at atherosclerotic cardiovascular disease, but it isn’t the dominant cause of disease [in your country], you’re going to be treating a whole host of people to try and tackle this thing that affects very few in a sense,” he says.
“It was not anything about REPRIEVE, it was a wonderful study, the hypothesis was tested, and it was shown to be correct, the intervention we know works,” Ntsekhe says. “It really then comes down to regional areas to think very carefully about how best they’re going to get their biggest bang for their buck,” he says. “We have to carefully consider the local context, local burden, we have set local health priorities, and weigh benefit and cost before we adopt new interventions or recommendations.”
SA’s cardiovascular disease burden
While Strijdom says we don’t have great data, he points to a large systematic review and meta-analysis published in 2018 in Circulation, which estimates that around 15% of the total cardiovascular disease burden in South Africa is attributable to HIV. “It’s probably higher than that. I would say that probably about one in five people with heart disease have heart disease because of HIV in South Africa,” he says, adding “that figure is probably only going to increase”.
Because of this, he says, there is a need for proper and clear primary healthcare guidelines specifically aimed at managing cardiovascular disease in people living with HIV, which we don’t currently have.
Strijdom says what we have at the moment since the rollout of the 2019 National ART Clinical Guidelines is very basic guidelines. This involves screening someone who has just been diagnosed with HIV by taking their blood pressure, and testing urine for glucose and proteins, and an assessment of their general cardiovascular disease risk by taking their medical and family history. These guidelines, according to Strijdom, only make provision for routine screening at baseline, but screening guidelines at follow-up visits are insufficient.
“I am, however, aware of the fact that there is progress especially from the integrated chronic disease management model which is currently being piloted in South Africa – and hopefully with that will come much more definitive and universal guidelines,” he says. “The bottom line is that South Africa, in its public health [sector] especially, really very quickly needs to come up with very clear and more comprehensive guidelines to actively manage cardiovascular disease risk in people with HIV.”
Need for annual screening
Strijdom suggest that to improve screening for cardiovascular disease risk in people living with HIV, there needs to be annual screening of people’s weight, their measure of body fat based on height and weight, waist circumference, blood pressure, cholesterol and triglyceride levels as well as testing urine samples for kidney function. There also needs to be a thorough family and medical history conducted for each patient.
“It’s not really a very expensive or very exhaustive list of stuff that you have to do. Unless of course they have specific symptoms and signs that leads you in a specific direction that you then have to perhaps do an ECG [a test used to evaluate the functioning of the heart] or cardiac imaging but that is usually determined by what you get from their history and clinical examination,” he says.
Ntsekhe says public health strategies to combat the growing burden of non-communicable diseases (NCDs), including cardiovascular disease, in South Africa must be strengthened. These include screening and prevention tools like checking a patient’s blood pressure and blood glucose, advising against smoking and alcohol as well as promoting health lifestyle choices like exercise and weight loss. These interventions should be offered to everyone, regardless of whether they are living with HIV or not, he says.
“The thing about NCDs and cardiovascular disease, for the most part, they are diseases of lifestyle and behaviour. So, when you talk prevention, it’s not always about drug prevention,” he says. “It’s more about intensification of those [interventions] that are already in the public domain, are very effective, and cost very little. Many of the public health and primary healthcare guidelines do advise local ministries, local health authorities on what should be happening.”
In terms of public education, Stritjdom says people need to be aware that there is something like high blood pressure. “If people are aware they will come to the clinic and will say please measure my blood pressure,” he says.
“Our health system is understandably focused on infectious diseases, but if we are not careful, we will then be totally unprepared to tackle the epidemic that will have replaced it. Namely, cancer, heart disease, stroke, obesity, diabetes, and it will totally overwhelm our public healthcare system,” he says.
Neurons in the brain of an Alzheimer’s patient, with plaques caused by tau proteins. Credit: NIH
A new study by investigators from Mass General Brigham suggests that whether a person inherits risk of Alzheimer’s disease from their mother or father influences risk of biological changes in the brain that lead to disease. By evaluating 4400 cognitively unimpaired adults ages 65–85, the team found those with a history of Alzheimer’s disease (AD) on either their mother’s side or both parents’ sides had increased amyloid in their brains. Their results are published in JAMA Neurology.
“Our study found if participants had a family history on their mother’s side, a higher amyloid level was observed,” said senior corresponding author Hyun-Sik Yang, MD, a neurologist at Mass General Brigham.
Yang said that previous smaller studies have investigated the role family history plays in Alzheimer’s disease. Some of those studies suggested maternal history represented a higher risk of developing Alzheimer’s, but the group wanted to revisit the question with cognitively normal participants and access to a larger clinical trial data set.
The team examined the family history of older adults from the Anti-Amyloid Treatment in Asymptomatic Alzheimer’s (A4) study, a randomized clinical trial aimed at AD prevention. Participants were asked about memory loss symptom onset of their parents. Researchers also asked if their parents were ever formally diagnosed or if there was autopsy confirmation of Alzheimer’s disease.
“Some people decide not to pursue a formal diagnosis and attribute memory loss to age, so we focused on a memory loss and dementia phenotype,” Yang said.
Researchers then compared those answers and measured amyloid in participants. They found maternal history of memory impairment at all ages and paternal history of early-onset memory impairment was associated with higher amyloid levels in the asymptomatic study participants. Researchers observed that having only a paternal history of late-onset memory impairment was not associated with higher amyloid levels.
“If your father had early onset symptoms, that is associated with elevated levels in the offspring,” said Mabel Seto, PhD, first author and a postdoctoral research fellow in the Department of Neurology at the Brigham. “However, it doesn’t matter when your mother started developing symptoms – if she did at all, it’s associated with elevated amyloid.”
Seto works on other projects related to sex differences in neurology. She said the results of the study are fascinating because Alzheimer’s tends to be more prevalent in women. “It’s really interesting from a genetic perspective to see one sex contributing something the other sex isn’t,” Seto said. She also noted the findings were not affected by whether study participants were biologically male or female.
Yang noted one limitation of the study is some participants’ parents died young, before they could potentially develop symptoms of cognitive impairment. He said social factors like access to resources and education may have also played a role in when someone acknowledged cognitive impairment and if they were ever formally diagnosed.
“It’s also important to note a majority of these participants are non-Hispanic white,” Seto added. “We might not see the same effect in other races and ethnicities.”
Seto said the next steps are to expand the study to look at other groups and examine how parental history affects cognitive decline and amyloid accumulation over time and why DNA from the mother plays a role.
Reisa Sperling, MD, a co-author on the paper, principal investigator of the A4 Study and a neurologist at Mass General Brigham, said the findings could be used soon in clinical translation.
“This work indicates that maternal inheritance of Alzheimer’s disease may be an important factor in identifying asymptomatic individuals for ongoing and future prevention trials,” Sperling said.
When a magnet is heated up, it reaches a critical point where it becomes demagnetisated. Called “criticality,” this point of high complexity is reached when a physical object is transitioning smoothly from one phase into the next.
Now, a new Northwestern University study has discovered that the brain’s structural features reside in the vicinity of a similar critical point – either at or close to a structural phase transition. Surprisingly, these results are consistent across brains from humans, mice and fruit flies, which suggests the finding might be universal. Although the researchers don’t know what phases the brain’s structure is transitioning between, they say this new information could enable new designs for computational models of the brain’s complexity and emergent phenomena.
“The human brain is one of the most complex systems known, and many properties of the details governing its structure are not yet understood,” said Northwestern’s István Kovács, the study’s senior author. “Several other researchers have studied brain criticality in terms of neuron dynamics. But we are looking at criticality at the structural level in order to ultimately understand how this underpins the complexity of brain dynamics. That has been a missing piece for how we think about the brain’s complexity. Unlike in a computer where any software can run on the same hardware, in the brain the dynamics and the hardware are strongly related.”
“The structure of the brain at the cellular level appears to be near a phase transition,” said Northwestern’s Helen Ansell, the paper’s first author. “An everyday example of this is when ice melts into water. It’s still water molecules, but they are undergoing a transition from solid to liquid. We certainly are not saying that the brain is near melting. In fact, we don’t have a way of knowing what two phases the brain could be transitioning between. Because if it were on either side of the critical point, it wouldn’t be a brain.”
While researchers have long studied brain dynamics using functional magnetic resonance imaging (fMRI) and electroencephalograms (EEG), advances in neuroscience have only recently provided massive datasets for the brain’s cellular structure. These data opened possibilities for Kovács and his team to apply statistical physics techniques to measure the physical structure of neurons.
For the new study, Kovács and Ansell analysed publicly available data from 3D brain reconstructions from humans, fruit flies and mice. By examining the brain at nanoscale resolution, the researchers found the samples showcased hallmarks of physical properties associated with criticality.
One such property is the well-known, fractal-like structure of neurons. This nontrivial fractal-dimension is an example of a set of observables, called “critical exponents,” that emerge when a system is close to a phase transition.
Brain cells are arranged in a fractal-like statistical pattern at different scales. When zoomed in, the fractal shapes are “self-similar,” meaning that smaller parts of the sample resemble the whole sample. The sizes of various neuron segments observed also are diverse, which provides another clue. According to Kovács, self-similarity, long-range correlations and broad size distributions are all signatures of a critical state, where features are neither too organised nor too random. These observations lead to a set of critical exponents that characterise these structural features.
“These are things we see in all critical systems in physics,” Kovács said. “It seems the brain is in a delicate balance between two phases.”
Kovács and Ansell were amazed to find that all brain samples studied – from humans, mice and fruit flies – have consistent critical exponents across organisms, meaning they share the same quantitative features of criticality. The underlying, compatible structures among organisms hint that a universal governing principle might be at play. Their new findings potentially could help explain why brains from different creatures share some of the same fundamental principles.
“Initially, these structures look quite different – a whole fly brain is roughly the size of a small human neuron,” Ansell said. “But then we found emerging properties that are surprisingly similar.”
“Among the many characteristics that are very different across organisms, we relied on the suggestions of statistical physics to check which measures are potentially universal, such as critical exponents. Indeed, those are consistent across organisms,” Kovács said. “As an even deeper sign of criticality, the obtained critical exponents are not independent – from any three, we can calculate the rest, as dictated by statistical physics. This finding opens the way to formulating simple physical models to capture statistical patterns of the brain structure. Such models are useful inputs for dynamical brain models and can be inspirational for artificial neural network architectures.”
Next, the researchers plan to apply their techniques to emerging new datasets, including larger sections of the brain and more organisms. They aim to find if the universality will still apply.
Human colon cancer cells. Credit: National Cancer Institute
Results from a new trial indicate that immunotherapy could successfully be used to treat the most common form of colorectal cancer, also known as bowel cancer.
The findings of the new study, a phase 1 trial involving the immunotherapy drugs botensilimab and balstilimab, have been published in the journal Nature Medicine, and it is the first time that consistent and durable responses to immunotherapy have been reported in difficult-to-treat patients.
Co-authored by Professor Justin Stebbing of Anglia Ruskin University (ARU), who describes the results as “potentially game changing”, the study focused on the most common type of colorectal tumours, known as MSS mCRC, or microsatellite stable metastatic colorectal cancer.
Although immunotherapy has previously been shown to work on patients with specific mismatch repair deficient (dMMR) tumours, only a small percentage of colorectal cancer patients have this type of tumour, and immunotherapy has so far been ineffective in patients with more common MSS mCRC tumours.
The new study involved using the immunotherapy drug botensilimab in conjunction with balstilimab on a group of patients in the United States. These drugs are both monoclonal antibodies, which work by triggering the body’s immune system to attack the cancer.
Of the patients in the phase 1 trial, 101 took part in a six-month follow-up and of these, 61% of them saw their tumour shrink or remain stable after receiving a combination of botensilimab (BOT) and balstilimab (BAL). The most common side-effects, or treatment-related adverse events, were diarrhoea and fatigue.
Justin Stebbing, Professor of Biomedical Sciences at Anglia Ruskin University (ARU) and communicating author of the study, said:
“These results are incredibly exciting. Colorectal or bowel cancer is one of the most common forms of cancer worldwide and this is the first time there has been convincing evidence that immunotherapy can work in all forms of colorectal tumours, so this is potentially game changing.
“This is now progressing into later phase clinical trials and we hope the FDA in the United States approve its use very soon. And because this is such an important area, affecting so many people, we hope authorities in the UK are also able to move quickly.” Joint first author Dr Andrea Bullock, Assistant Professor in Medicine at Beth Israel Deaconess Medical Center, said:
“This study sheds light on the potential of the BOT/BAL combination to treat microsatellite stable metastatic colorectal cancer, the most common form of colorectal cancer which has historically not responded to immunotherapy, and we hope our results will offer new hope for those diagnosed.” Joint last author Dr Anthony El-Khoueiry, Associate Director of Clinical Research and Chief of Section of Developmental Therapeutics at the USC Norris Comprehensive Cancer Center, said:
“This phase 1 study of botensilimab highlights its promising anti-tumour activity that encompasses immunologically cold tumours such as MSS colorectal cancer. The efficacy noted highlights the potential of botensilimab through its broader engagement of anti-tumour immunity.”
The full open access paper, published in , is available here
