User login
Risk models that incorporate genetic and circulating biomarker data in addition to established risk factors may improve risk modeling for pancreatic cancer in the general population, according to investigators.
Identifying high-risk individuals could facilitate earlier disease detection, which is essential for curing pancreatic cancer, reported lead author Jihye Kim, PhD, of Harvard School of Public Health, Boston, and colleagues.
“Given the late stage at presentation for most patients with pancreatic cancer, earlier detection approaches are worthy of significant investment as a critical means to reduce mortality from pancreatic cancer, soon to be the second-leading cause of cancer death in the United States,” the investigators wrote in Cancer Epidemiology, Biomarkers & Prevention.
According to the investigators, a variety of risk factors for pancreatic cancer are well established and include clinical, demographic, and lifestyle factors, while recent studies have reported associations with genetic and circulating biomarkers.
“Although risk factors have been investigated individually, their joint contribution to risk discrimination remains largely unknown,” the investigators wrote.
To learn more, Dr. Kim and colleagues performed a nested case-control study in which 500 patients with primary pancreatic adenocarcinoma were matched with 1,091 healthy controls. Data were drawn from four prospective studies: the Nurses’ Health Study, the Health Professionals Follow-up Study, the Women’s Health Initiative Observational Study, and the Physicians’ Health Study I. Via these studies, cases provided blood samples prior to diagnosis with pancreatic cancer.
“Importantly, because all our subjects were enrolled in prospective cohorts, all risk factor data and circulating markers were measured before the cases’ diagnosis of pancreatic cancer,” the investigators wrote. “This design faithfully recapitulates the situation faced by primary care physicians, where decisions related to disease screening are made in the prediagnostic setting using data collected in the several years prior to cancer diagnosis.”
In the present study, the investigators collected patient data for a variety of risk factors, including clinical and lifestyle characteristics, circulating biomarkers such as interleukin-6 and proinsulin, and 22 single-nucleotide polymorphisms. Frequencies and distributions of these factors were used to develop three multivariate risk models: a clinical model, a clinical/genetic model, and a clinical/genetic/biomarker model. To determine absolute risk of pancreatic cancer, these three models were combined with U.S. epidemiologic data, including incidence and mortality rates.
Cross-validation showed that the risk models became increasingly accurate with each added dataset; the area under the curve increased from 0.55 for the clinical model to 0.61 for the clinical/genetic model and ultimately to 0.62 for the clinical/genetic/biomarker model. Consequently, each model identified a greater number of individuals with at least a threefold risk of pancreatic cancer over a 10-year period. For example, the clinical model identified 1.5% of women and 0.2% of men with at least threefold risk, whereas the model that also included genetic and biomarker data identified 2.6% of women and 3.7% of men.
Absolute risk modeling allowed for generation of risk stratification percentiles by age. Women in the 99th risk percentile had a 1.7% risk of developing pancreatic cancer by age 70 years, and a 3.6% risk by age 80. For men, the highest-risk group had a 2.0% risk of pancreatic cancer by age 70 years and a 3.8% risk by age 80. Conversely, both men and women in the 10th risk percentile had a 0.2% risk by age 70 years and a 0.4% risk by age 80.
“[T]he addition of genetic variants and circulating markers added discriminatory ability beyond clinical factors that could be solicited in a physician’s office,” the investigators wrote.
“Further refinement and validation in independent samples will be necessary to make these models clinically actionable and impact survival of patients with pancreatic cancer,” they concluded.
The study was supported by the National Institutes of Health. The investigators reported additional relationships with Bayer, Celgene, Eli Lilly, and others.
SOURCE: Kim J et al. Cancer Epidemiol Biomarkers Prev. 2020 Apr 22. doi: 10.1158/1055-9965.EPI-19-1389.
Risk models that incorporate genetic and circulating biomarker data in addition to established risk factors may improve risk modeling for pancreatic cancer in the general population, according to investigators.
Identifying high-risk individuals could facilitate earlier disease detection, which is essential for curing pancreatic cancer, reported lead author Jihye Kim, PhD, of Harvard School of Public Health, Boston, and colleagues.
“Given the late stage at presentation for most patients with pancreatic cancer, earlier detection approaches are worthy of significant investment as a critical means to reduce mortality from pancreatic cancer, soon to be the second-leading cause of cancer death in the United States,” the investigators wrote in Cancer Epidemiology, Biomarkers & Prevention.
According to the investigators, a variety of risk factors for pancreatic cancer are well established and include clinical, demographic, and lifestyle factors, while recent studies have reported associations with genetic and circulating biomarkers.
“Although risk factors have been investigated individually, their joint contribution to risk discrimination remains largely unknown,” the investigators wrote.
To learn more, Dr. Kim and colleagues performed a nested case-control study in which 500 patients with primary pancreatic adenocarcinoma were matched with 1,091 healthy controls. Data were drawn from four prospective studies: the Nurses’ Health Study, the Health Professionals Follow-up Study, the Women’s Health Initiative Observational Study, and the Physicians’ Health Study I. Via these studies, cases provided blood samples prior to diagnosis with pancreatic cancer.
“Importantly, because all our subjects were enrolled in prospective cohorts, all risk factor data and circulating markers were measured before the cases’ diagnosis of pancreatic cancer,” the investigators wrote. “This design faithfully recapitulates the situation faced by primary care physicians, where decisions related to disease screening are made in the prediagnostic setting using data collected in the several years prior to cancer diagnosis.”
In the present study, the investigators collected patient data for a variety of risk factors, including clinical and lifestyle characteristics, circulating biomarkers such as interleukin-6 and proinsulin, and 22 single-nucleotide polymorphisms. Frequencies and distributions of these factors were used to develop three multivariate risk models: a clinical model, a clinical/genetic model, and a clinical/genetic/biomarker model. To determine absolute risk of pancreatic cancer, these three models were combined with U.S. epidemiologic data, including incidence and mortality rates.
Cross-validation showed that the risk models became increasingly accurate with each added dataset; the area under the curve increased from 0.55 for the clinical model to 0.61 for the clinical/genetic model and ultimately to 0.62 for the clinical/genetic/biomarker model. Consequently, each model identified a greater number of individuals with at least a threefold risk of pancreatic cancer over a 10-year period. For example, the clinical model identified 1.5% of women and 0.2% of men with at least threefold risk, whereas the model that also included genetic and biomarker data identified 2.6% of women and 3.7% of men.
Absolute risk modeling allowed for generation of risk stratification percentiles by age. Women in the 99th risk percentile had a 1.7% risk of developing pancreatic cancer by age 70 years, and a 3.6% risk by age 80. For men, the highest-risk group had a 2.0% risk of pancreatic cancer by age 70 years and a 3.8% risk by age 80. Conversely, both men and women in the 10th risk percentile had a 0.2% risk by age 70 years and a 0.4% risk by age 80.
“[T]he addition of genetic variants and circulating markers added discriminatory ability beyond clinical factors that could be solicited in a physician’s office,” the investigators wrote.
“Further refinement and validation in independent samples will be necessary to make these models clinically actionable and impact survival of patients with pancreatic cancer,” they concluded.
The study was supported by the National Institutes of Health. The investigators reported additional relationships with Bayer, Celgene, Eli Lilly, and others.
SOURCE: Kim J et al. Cancer Epidemiol Biomarkers Prev. 2020 Apr 22. doi: 10.1158/1055-9965.EPI-19-1389.
Risk models that incorporate genetic and circulating biomarker data in addition to established risk factors may improve risk modeling for pancreatic cancer in the general population, according to investigators.
Identifying high-risk individuals could facilitate earlier disease detection, which is essential for curing pancreatic cancer, reported lead author Jihye Kim, PhD, of Harvard School of Public Health, Boston, and colleagues.
“Given the late stage at presentation for most patients with pancreatic cancer, earlier detection approaches are worthy of significant investment as a critical means to reduce mortality from pancreatic cancer, soon to be the second-leading cause of cancer death in the United States,” the investigators wrote in Cancer Epidemiology, Biomarkers & Prevention.
According to the investigators, a variety of risk factors for pancreatic cancer are well established and include clinical, demographic, and lifestyle factors, while recent studies have reported associations with genetic and circulating biomarkers.
“Although risk factors have been investigated individually, their joint contribution to risk discrimination remains largely unknown,” the investigators wrote.
To learn more, Dr. Kim and colleagues performed a nested case-control study in which 500 patients with primary pancreatic adenocarcinoma were matched with 1,091 healthy controls. Data were drawn from four prospective studies: the Nurses’ Health Study, the Health Professionals Follow-up Study, the Women’s Health Initiative Observational Study, and the Physicians’ Health Study I. Via these studies, cases provided blood samples prior to diagnosis with pancreatic cancer.
“Importantly, because all our subjects were enrolled in prospective cohorts, all risk factor data and circulating markers were measured before the cases’ diagnosis of pancreatic cancer,” the investigators wrote. “This design faithfully recapitulates the situation faced by primary care physicians, where decisions related to disease screening are made in the prediagnostic setting using data collected in the several years prior to cancer diagnosis.”
In the present study, the investigators collected patient data for a variety of risk factors, including clinical and lifestyle characteristics, circulating biomarkers such as interleukin-6 and proinsulin, and 22 single-nucleotide polymorphisms. Frequencies and distributions of these factors were used to develop three multivariate risk models: a clinical model, a clinical/genetic model, and a clinical/genetic/biomarker model. To determine absolute risk of pancreatic cancer, these three models were combined with U.S. epidemiologic data, including incidence and mortality rates.
Cross-validation showed that the risk models became increasingly accurate with each added dataset; the area under the curve increased from 0.55 for the clinical model to 0.61 for the clinical/genetic model and ultimately to 0.62 for the clinical/genetic/biomarker model. Consequently, each model identified a greater number of individuals with at least a threefold risk of pancreatic cancer over a 10-year period. For example, the clinical model identified 1.5% of women and 0.2% of men with at least threefold risk, whereas the model that also included genetic and biomarker data identified 2.6% of women and 3.7% of men.
Absolute risk modeling allowed for generation of risk stratification percentiles by age. Women in the 99th risk percentile had a 1.7% risk of developing pancreatic cancer by age 70 years, and a 3.6% risk by age 80. For men, the highest-risk group had a 2.0% risk of pancreatic cancer by age 70 years and a 3.8% risk by age 80. Conversely, both men and women in the 10th risk percentile had a 0.2% risk by age 70 years and a 0.4% risk by age 80.
“[T]he addition of genetic variants and circulating markers added discriminatory ability beyond clinical factors that could be solicited in a physician’s office,” the investigators wrote.
“Further refinement and validation in independent samples will be necessary to make these models clinically actionable and impact survival of patients with pancreatic cancer,” they concluded.
The study was supported by the National Institutes of Health. The investigators reported additional relationships with Bayer, Celgene, Eli Lilly, and others.
SOURCE: Kim J et al. Cancer Epidemiol Biomarkers Prev. 2020 Apr 22. doi: 10.1158/1055-9965.EPI-19-1389.
FROM CANCER EPIDEMIOLOGY, BIOMARKERS & PREVENTION