FDA panel says blood test for colorectal cancer screening is safe, not effective

GAITHERSBURG, MD. – The Food and Drug Administration advisory panel deemed the first-of-a-kind blood test to screen for colorectal cancer safe but not effective.*

At a March 26 meeting, the FDA Molecular and Clinical Genetics Panel of the Medical Devices advisory committee* voted 9-0 with one abstention that the Epi proColon test, an in vitro diagnostic blood test manufactured by Epigenetics Inc., is safe to screen for colorectal cancer in patients with an average risk for the disease. Patients with a positive test would then be referred for diagnostic colonoscopy.

The panel was split in its view of the test’s effectiveness, however, voting 5-6 (the panel chair voted to break a 5-5 tie) that there was not a reasonable assurance of effectiveness for its proposed indications.

But it voted 5-4 (one abstention) that the benefits of the Epi proColon test outweighed its risks.

Epigenomics submitted its premarket approval application for the Epi proColon test last January for the detection of methylated Septim9 DNA in plasma derived from patient whole blood specimens. It currently has CE mark approval, and a first-generation version has been sold in Europe and the Middle East since 2009.

Two clinical studies were conducted to assess the clinical performance of Epi proColon. The pivotal clinical study was a prospective, multicenter trial comparing the test to colonoscopy using 6,857 samples collected in a previous noninferiority study. A supplemental clinical study was conducted comparing the performance of the Epi proColon test and a commercially available fecal immunochemical test (FIT) to colonoscopy.

In the pivotal trial, the Epi proColon test demonstrated a sensitivity of 68%, meeting the trial’s sensitivity target of 65%, but the test’s specificity of 80% did not meet the specificity target of 85%. The false-positive rate in the study was 21% higher than expected, indicating that more individuals than expected would be recommended to do a follow-up diagnostic procedure, such as a colonoscopy, which will be negative, the FDA said in its panel summary documents.

Some panelists questioned the sensitivity of the test, saying it was too low for a screening test, which should ideally have a sensitivity of 80%-90%. They also expressed concerns about the high false-positive rate.

"A sensitivity of about 70%, such that 30 out of every 100 people screened would be falsely negative, just seems too low as a screening test," said panelist Dr. Karen Weck, director of the molecular genetics laboratory at the University of North Carolina, Chapel Hill.

"The sponsor indicated that those individuals would still be recommended to have colonoscopy. So colonoscopy would still be recommended for individuals who test positive for this test and for individuals who test negative for this test, so it’s unclear to me what really is the clinical utility of this test then."

In the supplemental study, Epi proColon demonstrated noninferiority to the FIT for sensitivity but not for specificity, indicating that Epi proColon test exhibited a higher rate of false-positive results, compared with the FIT.

The pivotal trial also found that the false-positive rate increased with age and in African American subjects. Although panelists generally felt it unnecessary for the sponsor to include this information in additional labeling or warnings, several said that the false-positivity rates should somehow be conveyed to physicians and patients.

The panel supported the inclusion of the following warning statements in labeling with the test: not intended to replace colorectal cancer screening by colonoscopy; positive results are not confirmatory evidence for the presence of colorectal cancer; and negative results do not guarantee absence of cancer.

The sponsor and the FDA are working on a postapproval study for the test, with a possible study design of three annual test cycles and two additional years of study.

*Correction, 3/27/2014: An earlier version of this story did not make it clear that an advisory panel to the FDA made this recommendation.

GAITHERSBURG, MD. – The Food and Drug Administration advisory panel deemed the first-of-a-kind blood test to screen for colorectal cancer safe but not effective.*

At a March 26 meeting, the FDA Molecular and Clinical Genetics Panel of the Medical Devices advisory committee* voted 9-0 with one abstention that the Epi proColon test, an in vitro diagnostic blood test manufactured by Epigenetics Inc., is safe to screen for colorectal cancer in patients with an average risk for the disease. Patients with a positive test would then be referred for diagnostic colonoscopy.

The panel was split in its view of the test’s effectiveness, however, voting 5-6 (the panel chair voted to break a 5-5 tie) that there was not a reasonable assurance of effectiveness for its proposed indications.

But it voted 5-4 (one abstention) that the benefits of the Epi proColon test outweighed its risks.

Epigenomics submitted its premarket approval application for the Epi proColon test last January for the detection of methylated Septim9 DNA in plasma derived from patient whole blood specimens. It currently has CE mark approval, and a first-generation version has been sold in Europe and the Middle East since 2009.

Two clinical studies were conducted to assess the clinical performance of Epi proColon. The pivotal clinical study was a prospective, multicenter trial comparing the test to colonoscopy using 6,857 samples collected in a previous noninferiority study. A supplemental clinical study was conducted comparing the performance of the Epi proColon test and a commercially available fecal immunochemical test (FIT) to colonoscopy.

In the pivotal trial, the Epi proColon test demonstrated a sensitivity of 68%, meeting the trial’s sensitivity target of 65%, but the test’s specificity of 80% did not meet the specificity target of 85%. The false-positive rate in the study was 21% higher than expected, indicating that more individuals than expected would be recommended to do a follow-up diagnostic procedure, such as a colonoscopy, which will be negative, the FDA said in its panel summary documents.

Some panelists questioned the sensitivity of the test, saying it was too low for a screening test, which should ideally have a sensitivity of 80%-90%. They also expressed concerns about the high false-positive rate.

"A sensitivity of about 70%, such that 30 out of every 100 people screened would be falsely negative, just seems too low as a screening test," said panelist Dr. Karen Weck, director of the molecular genetics laboratory at the University of North Carolina, Chapel Hill.

"The sponsor indicated that those individuals would still be recommended to have colonoscopy. So colonoscopy would still be recommended for individuals who test positive for this test and for individuals who test negative for this test, so it’s unclear to me what really is the clinical utility of this test then."

In the supplemental study, Epi proColon demonstrated noninferiority to the FIT for sensitivity but not for specificity, indicating that Epi proColon test exhibited a higher rate of false-positive results, compared with the FIT.

The pivotal trial also found that the false-positive rate increased with age and in African American subjects. Although panelists generally felt it unnecessary for the sponsor to include this information in additional labeling or warnings, several said that the false-positivity rates should somehow be conveyed to physicians and patients.

The panel supported the inclusion of the following warning statements in labeling with the test: not intended to replace colorectal cancer screening by colonoscopy; positive results are not confirmatory evidence for the presence of colorectal cancer; and negative results do not guarantee absence of cancer.

The sponsor and the FDA are working on a postapproval study for the test, with a possible study design of three annual test cycles and two additional years of study.

*Correction, 3/27/2014: An earlier version of this story did not make it clear that an advisory panel to the FDA made this recommendation.

GAITHERSBURG, MD. – The Food and Drug Administration advisory panel deemed the first-of-a-kind blood test to screen for colorectal cancer safe but not effective.*

At a March 26 meeting, the FDA Molecular and Clinical Genetics Panel of the Medical Devices advisory committee* voted 9-0 with one abstention that the Epi proColon test, an in vitro diagnostic blood test manufactured by Epigenetics Inc., is safe to screen for colorectal cancer in patients with an average risk for the disease. Patients with a positive test would then be referred for diagnostic colonoscopy.

The panel was split in its view of the test’s effectiveness, however, voting 5-6 (the panel chair voted to break a 5-5 tie) that there was not a reasonable assurance of effectiveness for its proposed indications.

But it voted 5-4 (one abstention) that the benefits of the Epi proColon test outweighed its risks.

Epigenomics submitted its premarket approval application for the Epi proColon test last January for the detection of methylated Septim9 DNA in plasma derived from patient whole blood specimens. It currently has CE mark approval, and a first-generation version has been sold in Europe and the Middle East since 2009.

Two clinical studies were conducted to assess the clinical performance of Epi proColon. The pivotal clinical study was a prospective, multicenter trial comparing the test to colonoscopy using 6,857 samples collected in a previous noninferiority study. A supplemental clinical study was conducted comparing the performance of the Epi proColon test and a commercially available fecal immunochemical test (FIT) to colonoscopy.

In the pivotal trial, the Epi proColon test demonstrated a sensitivity of 68%, meeting the trial’s sensitivity target of 65%, but the test’s specificity of 80% did not meet the specificity target of 85%. The false-positive rate in the study was 21% higher than expected, indicating that more individuals than expected would be recommended to do a follow-up diagnostic procedure, such as a colonoscopy, which will be negative, the FDA said in its panel summary documents.

Some panelists questioned the sensitivity of the test, saying it was too low for a screening test, which should ideally have a sensitivity of 80%-90%. They also expressed concerns about the high false-positive rate.

"A sensitivity of about 70%, such that 30 out of every 100 people screened would be falsely negative, just seems too low as a screening test," said panelist Dr. Karen Weck, director of the molecular genetics laboratory at the University of North Carolina, Chapel Hill.

"The sponsor indicated that those individuals would still be recommended to have colonoscopy. So colonoscopy would still be recommended for individuals who test positive for this test and for individuals who test negative for this test, so it’s unclear to me what really is the clinical utility of this test then."

In the supplemental study, Epi proColon demonstrated noninferiority to the FIT for sensitivity but not for specificity, indicating that Epi proColon test exhibited a higher rate of false-positive results, compared with the FIT.

The pivotal trial also found that the false-positive rate increased with age and in African American subjects. Although panelists generally felt it unnecessary for the sponsor to include this information in additional labeling or warnings, several said that the false-positivity rates should somehow be conveyed to physicians and patients.

The panel supported the inclusion of the following warning statements in labeling with the test: not intended to replace colorectal cancer screening by colonoscopy; positive results are not confirmatory evidence for the presence of colorectal cancer; and negative results do not guarantee absence of cancer.

The sponsor and the FDA are working on a postapproval study for the test, with a possible study design of three annual test cycles and two additional years of study.

*Correction, 3/27/2014: An earlier version of this story did not make it clear that an advisory panel to the FDA made this recommendation.