What Is a Heart Attack? Symptoms, Causes, Diagnosis, Treatment, and Prevention

Diagnostic Test Detects Ovarian Cancer With 93% Accuracy

Ovarian cancer – a "silent killer"

Ovarian cancer is often referred to as a "silent killer" due to the unfortunate fact that symptoms often arise once the disease has reached an advanced stage. By this point, effective treatment strategies can be limited. According to the Ovarian Cancer Research Alliance, the 5-year survival rate for patients diagnosed with stage I ovarian cancer is 89%; for stage IV, it's 20%.

"Clearly, there is a tremendous need for an accurate early diagnostic test for this insidious disease," Dr. John McDonald, professor emeritus in the school of biological sciences at the Georgia Tech Integrated Cancer Research Center (ICRC), said. McDonald is also the founding director of the ICRC.

Over the last three decades, there have been numerous efforts to create a highly accurate early-detection test for ovarian cancer, with limited success. That's largely because cancer development is a highly heterogeneous process. While two patients might ultimately be diagnosed with the same type of cancer, their cells and tissues might have undergone very different molecular journeys to reach that point of diagnosis.

"Because of this high-level molecular heterogeneity among patients, the identification of a single universal diagnostic biomarker of ovarian cancer has not been possible," McDonald said.

At the ICRC, McDonald and colleagues sought to identify and develop a machine learning-based classifier, which utilizes metabolic profiles of serum samples, to accurately identify people with ovarian cancer. The team's research is published in Gynecologic Oncology.

Continue reading below...

Metabolic profiles in cancer

In metabolomics studies, mass spectrometry (MS) can help to identify what metabolites are present in a sample – such as blood – by detecting their mass and charge signatures.

What are metabolic profiles?

Metabolic profiles are a large set of biochemical markers and measurements that provide insight into an individual's metabolic state. They might include information on the levels of circulating lipids, proteins, carbohydrates and other metabolites that can be harnessed to create a picture of an individual's health.

MS only gets you so far, though. Identifying the exact chemical makeup of individual metabolites requires more extensive characterization, and only a small fraction of blood metabolites in the human body have been characterized. It's not possible, therefore, to accurately pinpoint the molecular processes that underpin an individual's metabolic profile – at least, not right now.

Even so, the presence of specific metabolites in the blood, as detected by MS, can be harnessed in the development of machine-learning based predictive models. "Because end-point changes on the metabolic level are known to be reflective of underlying changes operating collectively on multiple molecular levels, we chose metabolic profiles as the backbone of our analysis," said Dongjo Ban, a graduate research assistant in the McDonald lab, and first author of the study.

"The set of human metabolites is a collective measure of the health of cells," said co-author Professor Jeffrey Skolnick, "and by not arbitrary choosing any subset in advance, one lets the artificial intelligence figure out which are the key players for a given individual."

Continue reading below...

Utilizing artificial intelligence to develop an early diagnostic test for ovarian cancer

To obtain the data to train their model, McDonald and colleagues collected serum samples from 431 ovarian cancer patients and 133 healthy women across 4 locations: Northside Hospital, Atlanta (10 early- and 142 late-stage cancer samples), Fox Chase Cancer Center Biosample Repository Facility, Philadelphia (51 early- and 68 late-stage cancer samples, 133 control samples), University of North Carolina Medical School, Chapel Hill (17 early-stage cancer samples) and Alberta Health Services, Alberta (23 early- and 120 late-stage cancer samples).

"To help ensure the quality of our metabolic data, individual normal and ovarian cancer patient samples were collected from four geographically divergent locations and analyzed using ultra-performance liquid chromatography coupled with tandem mass spectrometry (UPLC-MS/MS-positive and negative modes and each sample independently pre-processed through two columns), generating four distinct datasets," the researchers described.

They then used recursive feature eliminiation (RFE) coupled with repeated cross-validation (CV) to identify the most reliable metabolites from the datasets.

What is recursive feature eliminaton and cross-validation in machine learning?

RFE is a method used in machine learning for feature selection, i.E., selecting a subset of the most important features from a dataset of features. In this study, "features" are the metabolites. In RFE, a model is trained on a dataset, where it ranks features based on specific criteria, and eliminates the least important features. This process is repeated several times.

CV is another technique that helps researchers evaluate the performance of machine learning models. By coupling RFE and CV, researchers can enhance the reliability of model evaluation and optimize feature selection.

McDonald and colleagues developed a consensus classifier – a final model – by aggregating the results of five independent machine learning algorithms. "The probabilities assigned to individuals by the consensus model were utilized to create a background distribution of probabilities that a given sample was cancer or normal," the researchers explained.

Model distinguishes cancer from controls with 93% accuracy

The consensus classification model was able to distinguish cancer from control samples with 93% accuracy, according to the researchers.

"This personalized, probabilistic approach to cancer diagnostics is more clinically informative and accurate than traditional binary (yes/no) tests," McDonald said. "It represents a promising new direction in the early detection of ovarian cancer, and perhaps other cancers as well."

Want more breaking news?

Subscribe to Technology Networks' daily newsletter, delivering breaking science news straight to your inbox every day.

Subscribe for FREE

The model requires further refinements and analyses. Its accuracy in predicting women with ovarian cancer was "slightly greater" than its accuracy in predicting women without the disease, the researchers explained in the paper. Currently, they do not know why, though they suggested that it could be due to the model potentially detecting disease in women prior to clinical symptoms and diagnosis. "Time course studies are currently being instituted to test this hypothesis," they said.

Reference: Ban D, Housley SN, Matyunina LV, et al. A personalized probabilistic approach to ovarian cancer diagnostics. Gynecol Oncol. 2024;182:168-175. Doi: 10.1016/j.Ygyno.2023.12.030

This article is a rework of a press release issued by the Georgia Tech Integrated Cancer Research Center. Material has been edited for length and content.

Minority Populations Largely Unrepresented In Gynecologic Cancer Clinical Trials

In a cohort with more than half a million women with endometrial, ovarian, or cervical cancer, less than 1% were enrolled in a clinical trial, and more than 85% of enrolled women were White.

New research further revealed disparities in gynecologic cancer clinical trial enrollment, warranting increased efforts to improve representation in these trials.

The cohort study was published in JAMA Network Open and included 562,592 women who had endometrial or uterine, ovarian, or cervical cancer between 2004 and 2019, using data from the National Cancer Database (NCDB) and the Surveillance, Epidemiology, and End Results Program (SEER).

Of the cohort, the vast majority (78.7%) of patients were White, with 10% Black, 6.8% Hispanic, 3.3% Asian, 0.3% American Indian/Alaska Native, and 1453 Native Hawaiian/Pacific Islander patients, and 3121 were classified as other race and ethnicity. The overall mean (SD) age at diagnosis was 62.9 (11.3) years.

With 548 of these women being enrolled in a gynecologic cancer clinical trial, this means less than 1% were represented, and this subset was largely comprised of White women (85.8%). A multivariable-adjusted model revealed that, compared with White women, clinical trial enrollment was lower among Asian (OR, 0.44; 95% CI, 0.25-0.78), Black (OR, 0.70; 95% CI, 0.50-0.99), and Hispanic (OR, 0.53; 95% CI, 0.33-0.83) women, who made up 2.2%, 7.1%, and 3.8% of the clinical trial subset, respectively.

Clinical trialImage credit: wladimir1804 – stock.Adobe.Com

These differences were not deemed significantly different for American Indian/Alaska Native women (OR, 1.37; 95% CI, 0.43-4.36), Native Hawaiian/Pacific Islander women (OR, 0.86; 95% CI, 0.12-6.16), or women of "other" race (OR, 0.48; 95% CI, 0.12-1.92), as the proportions of these women were more similar between the whole cohort and the clinical trial subset than for other races.

Interestingly, of the 3 included gynecologic cancer types, most women in the overall cohort had uterine cancer (58.6%), followed by ovarian (28.9%) and cervical (12.6%) cancer. However, the vast majority of women in clinical trials had ovarian cancer (78.3%), followed by uterine (12.2%) then cervical (9.5%) cancer.

According to the authors, they found intriguing patterns after analyses comparing the race-specific clinical trial enrollment prevalence within the NCDB sample with the corresponding race-specific cancer prevalence in the broader US population with gynecologic cancer. Participation-to-prevalence ratios (PPRs) were computed by dividing the percentage of clinical trial participants specific to each race and ethnicity in the study sample by the corresponding percentage of those racial and ethnic groups in the SEER database. Compared with the US population, White women were adequately or overrepresented for all 3 cancer types (PPRs ≥ 1.1), and Black women were adequately or overrepresented for endometrial and cervical cancers (PPRs ≥ 1.1) but underrepresented for ovarian cancer (PPR ≤ 0.6). Meanwhile, Asian and Hispanic women were underrepresented among all 3 cancer types (PPRs ≤ 0.6).

"Together, these analyses provide novel information on the landscape of racial and ethnic disparities in gynecologic cancer treatment," the authors said.

Disparities in gynecologic oncology between Black and White populations specifically have been extensively studied, but research on other racial and ethnic groups is limited. The current study revealed that Asian and Hispanic women are underrepresented and have lower odds of clinical trial enrollment compared with White women, aligning with previous research indicating lower enrollment of women from these groups in precision oncology and cervical cancer clinical trials. Additionally, a review of trials sponsored by the National Cancer Institute found that Hispanic women—but not Asian women—were less likely to be enrolled in ovarian, uterine, or cervical cancer clinical trials. Due to the small number of patients from Native Hawaiian/Pacific Islander, American Indian/Alaska Native, and other racial groups, meaningful estimates for clinical trial enrollment odds or PPRs could not be provided.

Besides race and ethnicity, clinical trial enrollment was influenced by other factors, including the presence of comorbidities, which correlated with decreased odds of participation—again consistent with previous research. Traditionally, clinical trials have excluded individuals with medical comorbidities for safety reasons. However, in 2017 and 2021, the American Society for Clinical Oncology recommended widening eligibility criteria to enhance generalizability. Other factors associated with lower clinical trial enrollment included older age; residence in zip codes with higher income and lower educational attainment; living in urban, small, or medium-sized counties; and receiving treatment in the South, Midwest, or Pacific regions, compared with the Northeast. On the other hand, treatment at an academic or research program or an integrated network cancer program was linked to higher odds of clinical trial participation.

"Most of these associations were expected on the basis of prior literature; however, our finding that women living in areas with higher area-level income were less likely to participate in clinical trials was surprising," the authors said. "It is likely that area-level income also captures unmeasured neighborhood effects underlying this unexpected association. Future studies that also include individual-level income measures will be useful in contextualizing this association."

The study's limitations stem from the available data in the NCDB, which lacks crucial information on patient and oncologic characteristics, trial specifics, and contextual information on therapeutic areas. Unmeasured confounding is also a possibility, and the absence of details on the pathway to clinical trial enrollment for different racial and ethnic groups hindered the authors' ability to recommend specific interventions. Additionally, the exclusion of approximately 45% of the original sample due to missing values in key variables may have led to imprecise estimates, particularly for American Indian/Alaska Native and Native Hawaiian/Pacific Islander women, highlighting the need for future studies to address these disparities.

"Further work aimed at understanding the race-specific barriers and facilitators that impact enrollment of gynecologic oncology patients in clinical trials is imperative," the authors concluded. "Although we noted lower clinical trial enrollment in multiple minoritized groups, the pathways leading to these outcomes are likely diverse and will require targeted interventions."

Reference

Khadraoui W, Meade CE, Backes FJ, Felix AS. Racial and ethnic disparities in clinical trial enrollment among women with gynecologic cancer. JAMA Netw Open. 2023;6(12):e2346494. Doi:10.1001/jamanetworkopen.2023.46494

MRNA Therapeutic Successfully Combats Ovarian Cancer In Mice

Each year, several thousand women in Germany die from ovarian cancer. In many cases, the disease is only detected when it is very advanced and metastases have already formed -- usually in the intestines, abdomen or lymph nodes. At such a late stage, only 20 to 30 percent of all those affected survive the next five years. "Unfortunately, this situation has hardly changed at all over the past two decades," says Professor Klaus Strebhardt, Director of the Department of Molecular Gynecology and Obstetrics at University Hospital Frankfurt.

96 percent of all ovarian cancer (high-grade) patients share the same clinical picture: The tumor suppressor gene p53 has mutated and is now non-functional. The gene contains the building instructions for an important protein that normally recognizes damage in the genetic material (DNA) of each cell. It then prevents these abnormal cells from proliferating and activates repair mechanisms that rectify the damage. If this fails, it induces cell death. "In this way, p53 is very effective in preventing carcinogenesis," explains Strebhardt. "But when it is mutated, this protective mechanism is eradicated."

If a cell wants to produce a certain protein, it first makes a transcript of the gene containing the building instructions for it. Such transcripts are called mRNAs. In women with ovarian cancer, the p53 mRNAs are just as defective as the gene from which they were copied. "We produced an mRNA in the laboratory that contained the blueprint for a normal, non-mutated p53 protein," says Dr. Monika Raab from the Department of Molecular Gynecology and Obstetrics, who conducted many of the key experiments in the study. "We packed it into small lipid vesicles, known as liposomes, and then tested them first in cultures of various human cancer cell lines. The cells used the artificial mRNA to produce functional p53 protein."

In the next step, the scientists cultivated ovarian tumors -- organoids -- from patient cells sourced by the team led by Professor Sven Becker, Director of the Women's Clinic at University Hospital Frankfurt. After treatment with the artificial mRNA, the organoids shrank and began to die.

To test whether the artificial mRNA is also effective in organisms and can combat metastases in the abdomen, the researchers implanted human ovarian tumor cells into the ovaries of mice and injected the mRNA liposomes into the animals some time later. The result was very convincing, says Strebhardt: "With the help of the artificial mRNA, cells in the animals treated produced large quantities of the functional p53 protein, and as a result both the tumors in the ovaries and the metastases disappeared almost completely."

That the method was so successful is partly due to recent advances in mRNA technology: Normally, mRNA transcripts are very sensitive and degraded by cells within minutes. However, it is meanwhile possible to prevent this by specifically modifying the molecules. This extends their lifespan substantially, in this study to up to two weeks.

In addition, the chemical composition of the artificial mRNA is slightly different to that of its natural counterpart. This prevents the immune system from intervening after the molecule has been injected and from triggering inflammatory responses. In 2023, the Hungarian scientist Katalin Karikó and her American colleague Drew Weissman were awarded the Nobel Prize in Physiology or Medicine for this discovery. "Thanks to the development of mRNA vaccines such as those of BioNTech and Moderna, which went into action during the SARS-CoV-2 pandemic, we now also know how to make the molecules even more effective," explains Strebhardt.

Strebhardt, Raab and Becker are now looking for partners to join the next step of the translational project: testing on patients with ovarian cancer. "What is crucial now is the question of whether we can implement the concept and the results in clinical reality and use our method to help cancer patients," says Strebhardt. The latest results make him very optimistic that the tide could finally turn in the treatment of ovarian carcinomas. "p53 mRNA is not a normal therapeutic that targets a specific weak point in cancer cells. Instead, we are repairing a natural mechanism that the body normally uses very effectively to suppress carcinogenesis. This is a completely different quality of cancer therapy."

---