FDA Discusses How to Regulate NGS Tests for Cancer Diagnoses

The Food and Administration (FDA) Drugs for comments on how to regulate testing next-generation sequencing (NGS) that can analyze genes from a tumor to guide treatment of cancer.

In a public workshop on the campus of the FDA in Silver Spring, MD, the oncology experts NGS Thursday and discussed the challenges and considerations for establishing standards to validate what is known as oncology NGS panels or evidence based.

In recent years, the increasing understanding of genetics showed that genetic variation may have an effect on growth and the likelihood of a disease, and how a patient may respond to special treatment.

With this knowledge, drug manufacturers have worked to develop very specific therapies that rely on companion diagnostics to identify a gene or set of genes that could affect an effective treatment for a particular patient. So far, the FDA has approved more than 20 tests as companion diagnostics for cancer treatments.

Ten miles south of the FDA on Thursday, President Barack Obama addressed the importance of genetic variations in cancer at a White House summit on precision medicine.

"Sectors such as cancer show that each cancer may be unique -. Even if it is the same organ All these ideas promise the possibility of being able to cure diseases that until now we could not understand," Obama said.
Unlike traditional diagnosis, which usually detects the presence of one or more biomarkers or genes based tests can detect NGS "many biomarkers [and] Relevant mutations simultaneously."

Because of this, evidence-based NGS are very promising for oncology, but "bring a challenge to the diagnostic paradigm current accompaniment" said Reena Philips, director of the Center for Devices and Radiological Health (CDRH) division of molecular genetics and pathology.

Since 2014, the FDA has been working to develop a regulatory framework for NGS diagnosis based on validated analytical standards "that would ensure the NGS tests produce accurate and reliable results."

The objective of the workshop FDA, according to a working paper distributed by the agency, is to get "input on strategies to establish performance characteristics based panels oncology NGS" as well as concrete ideas on how to validate "rare variants across different types of tumors, monitoring of applications for companion diagnostics, and after the approval of changes dosing".

Preanalytical challenges and monitor quality metrics

One of the areas expected to get comments from the FDA on how different preanalytical factors is, as the tumor type / location and how a sample is prepared and can affect the performance of a diagnosis.

For example, if a basic diagnostic NGS require validation for different types of acquisition methods or tissue sample? Or there are "pre-analytical tests appropriate ... that can be used in place" to apply a test based on NGS?
John Pfeifer, professor of pathology and immunology at the University of Washington School of Medicine in St. Louis, said the choice of method used to prepare tissue samples is much less relevant to the diagnostic result that takes into account the fabric itself .

Because tumors have a high degree of heterogeneity, labs can get different results depending on the tumor sample.

"The importance of this is two laboratories can see the same fabric and differ as to what they found mutations in important driver genes, not because of differences in the quality of [its] analysis, while simply because they are looking at different parts of the tumor, "Pfeifer said. "There is no more variability in their analysis in terms of coverage based [guanine-cytosine] GC-content between different individual genes locations between formalin fixation or fresh tissue [frozen]," he added, referring to the two methods common for the preparation of tissue samples.

In response to a question about whether NGS-based diagnostics must be validated for each type of tissue (lung, liver, pancreas, etc.), Dara Aisner, a molecular genetic pathologist at the University of Colorado, asked: "In how is needed to validate the origin of the fabric separately? it's really different skin lung liver so different? ... Almost all the samples we see, with few exceptions, consists of the basic components and ... from my point of view, lungs and skin and liver are largely equivalent. "

Aisner suggested that the focus should be on tumors produce chemicals such as mucin or melanin, which can interfere with diagnostic results. He also warned that too many rules for methods of pre-analytical sample could limit access to NGS-based diagnostics and interfere with diagnostic practices.

"The mitigating factor is that there are mechanisms to evaluate the integrity of the nucleic acid ... for me the solution here is to establish measures to address the resulting product, namely, the integrity of the nucleic acid or NGS data," Aisner Said.

Rajyalakshmi Luthra, director of the laboratory of molecular diagnostic Cancer Center MD Anderson of the University of Texas, agrees, adding that setting limits too strict could cause laboratories reject samples for testing when they are below a certain threshold . "When we have strictly followed the [DNA] concentration ... cut in approximately 10% of cases, we would not even try, but when we cut the [threshold] DNA concentration, our success rate of 85% to 95 % ".

However, Pfeifer noted that certain standards or thresholds are necessary. "From a guy who is involved in a lab that does this, we have to seriously consider where there is minimal ... somehow we have to find a way we can get some clarity, some reproducibility," he said.

"Laboratories should be able to demonstrate that they can extract nucleic acids with high quality ... The question is, how sustainable is the search for these differences? Where is the line drawn?" Pfeifer asked. More importantly, he added, it is that the requirement regarding the methods of pre-analytical "not good if your pipe ... bioinformatic not granted appropriately."

Pan-Cancer Claims

While a diagnosis of pan NGS-based cancer remains elusive, companies pursue technology. In September, the FDA sent a letter to Pathway Genomics disputing his claim that the test can detect up to 10 types of cancer, and in January, Illumina announced the launch of a new company to develop a blood test that could detect many types of cancer "in the early stages."

In the workshop the FDA on Thursday, the agency was interested in practical issues that could help create standards to support diagnostic applications called "pan-cancer".

"What kind of tumor were the most difficult for you to get data from NGS, and what types of tumors that should be excluded from Pan-cancer claims unless the manufacturer specifically produced data of this type of tumor?" Aaron Schetter, a scientific reviewer for diagnosing in vitro CDRH office asked panelists.

Luthra said his group had reproducible NGS-based tests when they had resection of the tumor samples, but has been less successful with smaller tumors results.

"This is not the type of tumor, which is the specimen DNA ... and the way it is set, which is more important than the type of tumor," he said.

Labeling and Approval

Another issue raised Pfeifer, it is important that developers of evidence-based NGS to specify which types of mutations can indeed find their evidence. "It is absolutely clear that the doctor may order the differences in the various types of mutation that this test is designed to detect," he said.

Schetter said the FDA expects "to have labeling so that such limitations [are] present."

In the working document of the FDA, the agency offers two tables that may appear on the label of a test based NGS cancer. One that specifies the diagnostic indications complementary test is approved, and another that lists other genetic variations, the test can detect that have not validated to guide targeted therapy "in patients who have already been taken into account all appropriate therapies" .

Donna Roscoe, head of the branch of molecular genetic diagnostic in vitro Office CDRH and Radiological Health, stressed the importance of gathering data to support the use of a diagnostic, saying: "We manufacturers want us to go to market their onco [ology] forums [saying] "? what do you need us and tells us' what is its function"

He added that one thing will seek FDA is a defined protocol covering all stages of the results of tissue samples to demonstrate the performance of an expected diagnosis.

The Clinical Relevance

During the final discussion of the day, the FDA asked panelists that evidence based on evidence NGS must rely on clinical indications.

Abraham Tzou, a doctor at CDRH, asked: "What would be the type of performance measures ... to provide evidence of safety and effectiveness for monitoring escorts and say what types of design considerations diagnostic study is important ? "

"I'm the purpose of these tests, that user sees patients," said Dane Dickson, a general oncologist and medical director of the molecular evidence Development Consortium. Dickson expressed skepticism about the adoption of evidence-based NGS without demonstrating that evidence to improve patient outcomes.
"We need proof? Absolutely ... end of testing should be based on clinical," he said.

Apostolia Maria Tsimberidou, associate professor in the MD Anderson Cancer Center, agreed, saying teacher. "I think it's important ... make sure it is of clinical relevance need to standardize next-generation sequencing and set some rules ... before use them to make clinical decisions. "