Skip Navigation

Summary of Vol 102 Issue 4 Transcript

Hi and welcome to the JNCI podcast, a production of the Journal of the National Cancer Institute. I’m Steve Graff, JNCI’s assistant news and media editor, and I’ll be taking you through issue 4.

In this issue’s first article, researchers looked at the number of oncology and hematology drug reviews and subsequent approvals at the FDA from 2005 to 2007. Also in this issue, a study and an editorial focused on symptoms and their value for early detection of ovarian cancer, while another found that three commonly used esophageal cancer cell lines in research proved to be from other cancers. In another study, researchers found that the addition of the contrast agent gadolinium during MRI for primary tumor assessment improved accuracy for detecting lymph node metastases. And in this issue’s last study, researchers found that childhood survivors of CNS cancers and leukemia have lowered educational success.

All that in more detail coming up next…..

Over a two and half year period, beginning in 2005 when the U.S. Food and Drug Administration’s oncology drug product’s office began reviewing marketing applications, a total of 60 new oncology and hematology drugs were reviewed, of which 53 were approved, according to a new article published online January 29.

To provide an overview of recent regulatory actions by the FDA’s Office of Oncology Drug Products in the Center for Drug Evaluation and Research, Rajeshwari Sridhara, of the FDA’s Office of Biostatistics, in Silver Spring, Md., and colleagues identified all applications reviewed, as well as actions taken, from July 1, 2005, through December 31, 2007. Their review included “New Drug Application” and “Biologics Licensing Application” approvals.

Marketing applications for 60 new products were reviewed and regulatory action was taken on 58 of them based on a risk–benefit evaluation. Products that demonstrated efficacy and had an acceptable risk–benefit ratio (i.e., the magnitude of the treatment effect was statistically persuasive and clinically meaningful) were granted either regular or accelerated marketing approval. A total of 53 new indications were approved: 39 received regular approval, nine received accelerated approval, and five were converted from accelerated to regular approval.

According to an article published online January 28, use of symptoms to trigger a medical evaluation for ovarian cancer does not appear to detect early-stage ovarian cancer earlier and would likely result in diagnosis in only 1 out of 100 women in the general population with such symptoms.

Researchers at Fred Hutchinson Cancer Research Center in Seattle assessed the predictive value of certain symptoms, including abdominal pain or bloating and urinary frequency, which were cited in a recent consensus statement as a way to diagnose ovarian cancer earlier.

Mary Anne Rossing, of the Program in Epidemiology at Fred Hutchinson, and colleagues conducted in-person interviews with 812 patients aged 35-74 years who had epithelial ovarian cancer that was diagnosed from 2002 through 2005. They compared the results from these case patients with results from interviews with 1,313 population-based control subjects—women who did not have ovarian cancer. The researchers assessed the sensitivity, specificity, and positive predictive value of a proposed symptom index and of symptoms included in the consensus recommendation.

Symptoms appeared in most case patients only about 5 or fewer months before diagnosis. Women with early-stage ovarian cancer were somewhat less likely to have symptoms than those with late-stage cancer. The authors conclude that 100 symptomatic women would need to be evaluated to detect one woman with ovarian cancer.

In an accompanying editorial, Beth Y. Karlan, and Ilana Cass, at Cedars-Sinai Medical Center in Los Angeles, note the strengths of the study, including in-person interviews and large number of patients, but also point out its limitations, which includes inherent recall bias and survival bias in case patients and control subjects.

Three frequently used human esophageal adenocarcinoma cell lines used for research were confirmed as being from other tumor types, according to a brief communication published online January 14. Two of the cell lines have been used in 11 U.S. patents and more than 100 published studies.

The 13 established cell lines are important because of the limited availability of patient samples and animal models.

To determine the authenticity of all the available cell lines, Winand N.M. Dinjens, University Medical Center, in Rotterdam, the Netherlands, and colleagues used data from pathology archives and genotyping assays in collaboration with the primary investigators who established the cell lines.

Cell lines SEG-1, BIC-1, and SK-GT-5 were proven to be cell lines from other tumor types, including lung carcinoma, colorectal adenocarcinoma, and gastric fundus.

The researchers also suggest that the clinical trial involving Barrett-related esophageal adenocarcinoma patients on the drug sorafenib should be reconsidered since the wrong cell line was used. They say there is now scant evidence for inhibition of the mitogen-activated protein kinase pathway by the drug in this cancer.

In an accompanying editorial, Robert Shoemaker, of the NCI at Frederick in Maryland, questions this suggestion, pointing out that tissue of origin may not be important for all research studies. “…Given the knowledge that cancer is a heterogeneous disease,” he writes, “one might question the rationale for any therapeutic maneuver that is based on studies conducted on a single cell line.”

Shoemaker suggests that a study conducted with the correct cell lines (esophageal adenocarcinomas) would probably provide the same rationale for sorafenib because alterations in mitogen-activated protein kinase pathways are common in many tumor types.

Addition of the contrast agent gadolinium during magnetic resonance imaging for primary tumor assessment improved accuracy for detecting lymph node metastases, according to a new study published online February 1. Gadolinium-enhanced MRI is primarily used to visualize primary tumors and increasingly to detect and evaluate lymph node metastases.

Wenche M. Klerkx, at the University Medical Centre Utrecht, the Netherlands, and colleagues searched the literature for studies that compared the diagnostic accuracy of gadolinium-enhanced MRI for staging lymph node metastases with that of histopathologic examination. The researchers conducted a meta-analysis that encompassed more than 30 studies from the last 10 years and reported summary sensitivity and specificity of MRI for detecting nodal metastases.

The researchers found that overall accuracy of gadolinium-enhanced MRI for the detection of nodal metastases was moderate. They also concluded that incorporating contrast enhancement in the malignancy criteria improves the accuracy of this diagnostic test.

According to an article published online January 27, childhood cancer survivors who had brain or other central nervous system cancers, or leukemia, achieve lower-than-expected educational success compared with the general public.

E.R. Lancashire, of the Centre for Childhood Cancer Survivor Studies, School of Health and Population Sciences, at the University of Birmingham in the UK, and colleagues investigated educational attainment among a cohort of over 10,000 5-year survivors of all types of childhood cancer who had completed a British Childhood Cancer Survivor Study questionnaire. Educational attainment at two levels of secondary school (advanced and ordinary levels) up to the highest level of education was considered. Results were compared with data from the general population obtained from the General Household Survey of Great Britain.

Overall, survivors of childhood cancer experience a deficit in educational attainment compared with the general public. Specifically, only those who had central nervous system neoplasms, especially if they had received radiation to the head, and leukemia survivors receiving similar treatment, performed worse than the general population. These differences were substantial and statistically significant.

That’s it for this edition of the JNCI podcast. Check out this month’s JNCI interview with Dr. Joann Elmore, of the University of Washington, who discusses the JNCI commentary and article on DCIS.

And as always, find these studies online at or listen to more podcasts at

Let us know how we’re doing by emailing

For the latest cancer news and studies, also follow us on twitter @JNCI_Now

I’m Steve Graff, and thanks for listening…