Cancer therapies targeting specific molecular subtypes of the disease allow physicians to tailor treatment to a patient’s individual molecular profile. But scientists are finding that in many types of cancer the molecular subtypes are more varied than previously thought and contain further genetic alterations that can affect a patient’s response to therapy.
A UNC-led team of scientists has shown for the first time that lung cancer molecular subtypes correlate with distinct genetic alterations and with patient response to therapy. These findings in pre-clinical models and patient tumour samples build on their previous report of three molecular subtypes of non-small cell lung cancer and refines their molecular analysis of tumours.
Study senior author, Neil Hayes, MD, MPH, associate professor of medicine, says, ‘It has been known for about a decade of using gene expression arrays that ‘molecular subtypes’ exist. These subtypes have molecular ‘fingerprints’ and frequently have different clinical outcomes. However, the underlying etiologies of the subtypes have not been recognised. Why do tumours form subtypes?
‘Our study shows that tumour subtypes have different underlying alterations of DNA as part of the difference. These differences are further evidence of the importance of subtypes and the way we will use them. For example, the mutations are different which may imply much more ability to target than previously recognised. Also, we are starting to get a suggestion that these subtypes may reflect different cells of origin that rely on different cancer pathways. This is further unlocking the diversity of this complex disease.’ Hayes is a member of UNC Lineberger Comprehensive Cancer Center.
The team first defined and reported in 2006 on three lung cancer molecular subtypes, named according to their genetic pattern – bronchoid, squamoid and magnoid.
In this paper they sought to determine if distinct genetic mutations co-occur with each specific molecular subtypes. They found that specific genetic mutations were associated with each subtype and that these mutations may have independent predictive value for therapeutic response. Lineberger Clinical/Translational Developmental Research Award.

One of the largest studies to investigate birth complications and later mental health has found that premature birth constitutes a single, independent risk factor for a range of severe psychiatric disorders. Researchers at King’s College London in the UK and Karolinska Institutet suggest that neurodevelopmental differences in those born prematurely may be important in understanding the link.
Researchers identified all individuals registered in the Swedish birth register between 1973 and 1985 who were alive and living in Sweden at the age of 16, a total of nearly 1.5 million individuals. By analysing national hospital discharge registers, they identified all individuals admitted to hospital with their first episode of a psychiatric disorder.
The study found that individuals born extremely prematurely (less than 32 weeks gestation) were 2.5 times more likely to have psychosis as young adults, nearly 3 times more likely to have depression, and 7.4 times more likely to have bipolar disorder than those born at term (37-41 weeks gestation). The findings also revealed a smaller, yet still significant, increased risk of developing bipolar disorder, psychosis and depression for those born moderately prematurely (32-36 weeks).
The study also investigated the link between premature birth, eating disorders and alcohol and drug dependency, but association with these disorders was much weaker. Other adverse perinatal factors including newborn health, maternal socio-demographic characteristics and maternal psychiatric history were taken into account and were found to have no significant effect on the findings.
‘We believe that the increased risk of mental disorders in those born very prematurely can be explained by alterations of brain development’, says Professor Christina Hultman at Karolinska Institutet, who led the Swedish part of the study. ‘The immature nervous system in these children is particularly vulnerable to brain injury resulting from birth complications.’
Approximately 6 percent of babies in Sweden are born prematurely every year. Thanks to research and new technology, today many of prematurely born can be saved. Most of these babies go on to lead healthy lifestyles, although as a group they are more likely to require extra school support and be hospitalised with a variety of physical problems. Therefore the authors point to the importance of raising awareness of the increased risk of mental health disorders in people born prematurely, and suggest gestational age should be considered when investigating risk factors for psychiatric disorders in young adults. Karolinska Institute

The addition of changes in inflammatory biomarkers to established clinical variables improves the prediction of mortality in patients with chronic obstructive pulmonary disease (COPD), according to a new study.
‘COPD is characterised by low-grade inflammation, so we hypothesised that the addition of inflammatory biomarkers to established predictive factors would improve the prediction of mortality,’ said lead author Bartolome Celli, lecturer in medicine at Harvard Medical School and member of the Pulmonary and Critical Care Division of Brigham and Women’s Hospital in Boston. ‘We found that the addition of a panel of selected biomarkers to clinical variables significantly improved the ability of clinical variables to predict mortality in these patients.’
The researchers analysed prospectively collected data on 1,843 COPD patients from the Evaluation of COPD Longitudinally to Identify Predictive Surrogate Endpoints (ECLIPSE) study. Of these 1,843 patients, 168 (9.1%) died during the three-year follow-up.
Clinical predictors of morality included age, BODE (Body-Mass Index, Airflow Obstruction, Dyspnea, and Exercise Capacity) index , and incidence of hospitalisations due to exacerbation’s of COPD in the year prior to the study. A predictive model for mortality using these clinical variables had a C-statistic (which measures the ability of how well a clinical prediction rule can correctly rank-order patients by risk) of 0.686. Adding interleukin-6 (IL-6) to the predictive model significantly improved the C-statistic to 0.708, and the addition of a panel of biomarkers including white blood cell counts, IL-6, C-reactive protein (CRP), interleukin-8 (IL-8), fibrinogen, chemokine (C-C-motif) ligand 18 (CCL-18), and surfactant protein D (SP-D) further improved the C-statistic to 0.726.
‘This panel of selected biomarkers was not only elevated in non-survivors in our cohort, but was associated with mortality over three years of follow-up after adjusting for clinical variables known to predict mortality in patients with COPD,’ said Dr. Celli. ‘Except for IL-6, these biomarkers improved the predictive value of our model only marginally when considered individually, but they improved the model significantly when analyzed as a group.’
The study had several limitations, including the lack of a study adjudication committee to specify causes of death, the exclusion of some biomarkers thought to be important in the pathobiology of COPD, and the lack of a validating cohort.
‘Adding white blood cell counts and measurement of changes in systemic levels of IL-6, CRP, IL-8, fibrinogen, CCL-18, and SP-D significantly improves the ability of clinical variables to predict mortality in patients with COPD,’ said Dr. Celli. ‘This is the first study to show that the addition of biomarker levels to clinical predictors in COPD patients adds relevant prognostic information.’ EurekAlert

For 5,000 years, the Chinese have used a system of medicine based on the flow and balance of positive and negative energies in the body. In this system, the appearance of the tongue is one of the measures used to classify the overall physical status of the body, or zheng. Now, University of Missouri researchers have developed computer software that combines the ancient practices and modern medicine by providing an automated system for analysing images of the tongue.
‘Knowing your zheng classification can serve as a pre-screening tool and help with preventive medicine,’ said Dong Xu, chair of MU’s computer science department in the College of Engineering and study co-author. ‘Our software helps bridge Eastern and Western medicine, since an imbalance in zheng could serve as a warning to go see a doctor. Within a year, our ultimate goal is to create an application for smartphones that will allow anyone to take a photo of their tongue and learn the status of their zheng.’
The software analyses images based on the tongue’s colour and coating to distinguish between tongues showing signs of ‘hot’ or ‘cold’ zheng. Shades of red and yellow are associated with hot zheng, whereas a white coating on the tongue is a sign of cold zheng.
‘Hot and cold zheng doesn’t refer directly to body temperature,’ said Xu, who is also on the faculty of the Bond Life Sciences Center. ‘Rather, it refers to a suite of symptoms associated with the state of the body as a whole.’
For example, a person with cold zheng may feel chills and coolness in the limbs and show a pale flushing of face. Their voice may have a high pitch. Other symptoms of cold sheng are clear urine and loose stool. They also may prefer hot foods and drinks and desire warm environments.
In Chinese traditional medicine both hot and cold zheng can be symptoms of gastritis, an inflammation of the stomach lining frequently caused by bacterial infection.
For the study, 263 gastritis patients and 48 healthy volunteers had their tongues analysed. The gastritis patients were classified by whether they showed infection by a certain bacteria, known as Helicobacter pylori, as well as the intensity of their gastritis symptoms. In addition, most of the gastritis patients had been previously classified with either hot or cold zheng. This allowed the researchers to verify the accuracy of the software’s analysis.
‘Our software was able to classify people based on their zheng status,’ said study co-author Ye Duan, associate professor of computer science at MU.
‘As we continue to work on the software we hope to improve its ability,’ Duan said. ‘Eventually everyone will be able to use this tool at home using webcams or smartphone applications. That will allow them to monitor their zheng and get an early warning about possible ailments.’ University of Missouri

Researchers have identified a potential new pathway in prostate cancer cells by which cancer-driving gene products can be generated, according to a study.
‘Our work shows that cancers have many more tricks than we thought to generate potential cancer-driving genes or gene products,’ said Hui Li, Ph.D., assistant professor of pathology at the University of Virginia in Charlottesville, and a recipient of an Innovative Research Grant from Stand Up To Cancer (SU2C). The AACR is the scientific partner of SU2C.
Gene fusion is a common characteristic of human cancers. In many cases, the protein products of these gene fusions, which are generated via an RNA intermediate, have a key role in the genesis of the cancer. A well-characterised example of this is the protein that drives chronic myeloid leukaemia, BCR-ABL, which is generated via RNA intermediates from a fusion gene formed by chromosomal translocation — an event involving exchange of genomic DNA between two distinct chromosomes.
‘For many years, chromosomal translocation was considered the sole way in which single RNAs consisting of copies of parts of two genes, so-called fusion RNAs, could be generated,’ said Li. ‘We have shown that fusion RNAs can be generated without changes to DNA by a new mechanism that we are calling cis-SAGe [cis-splicing of adjacent genes].’ Recently, a fusion RNA formed from parts of the SLC45A3 and ELK4 genes was identified in prostate cancer cells in the absence of any DNA alterations. Li and his colleagues confirmed in two prostate cancer cells lines that the SLC45A3-ELK4 fusion RNA could be detected even though there was no evidence of genomic DNA rearrangement.
Detailed molecular analysis of the prostate cancer cell lines indicated that the SLC45A3-ELK4 fusion RNA was generated by cis-SAGe. SLC45A3 and ELK4 are neighbouring genes, and cis-SAGe occurred when an RNA that crossed the boundary between the two genes was formed.
The protein CCCTC-binding factor normally acts to insulate SLC45A3 and ELK4 from each other. Li and his colleagues found that levels of this protein at the gene boundary inversely correlated with the amount of SLC45A3-ELK4 fusion RNA generated, providing molecular insight into how the quantity of this fusion RNA could be regulated.
A functional role for the SLC45A3-ELK4 fusion RNA in prostate cancer was suggested by two observations. First, it promoted the growth of the two prostate cancer cell lines in culture. Second, its levels in human prostate samples correlated with prostate cancer disease progression — normal prostate tissue expressed the lowest levels and prostate cancer specimens from men with metastatic disease expressed the highest levels.
‘These data are not sufficient to say that the SLC45A3-ELK4 fusion RNA has a causal role in prostate cancer,’ said Li. ‘But they are highly suggestive, and I am very excited that this high-risk project, which I would not have been able to pursue without the grant from Stand Up To Cancer, has uncovered what seems to be a new way in which cancer can be driven.’ EurekAlert