Clinical Laboratory Strategies: February 10, 2011

Strategies logo

Using Tumor Enrichment Strategies for Mutation Detection
Fast COLD PCR Method shows promise for KRAS detection in colorectal cancer
By Genna Rollins



Robust molecular diagnostic methods have become crucial to the oncology field, where clinicians increasingly factor a patient’s specific genetic makeup into their clinical decision-making. This is particularly true in tumor mutation analysis, when specimens may have a small amount of mutant cells and the mutation-detection limit of traditional genetic sequencing techniques is about 20%. Consequently, labs have implemented numerous DNA enrichment methods, and this issue of Strategies reports on one.

The emergence of personalized medicine linking a patient’s individual genetic makeup with specific therapies is transforming medicine and placing added emphasis on the quality of molecular diagnostics. One of the best examples is KRAS mutation analysis in colorectal cancer, in which studies have demonstrated better treatment outcomes from anti-epidural growth factor receptor (EGFR) therapy in patients with KRAS wild-type status versus KRAS mutations. Based on these findings, the Food and Drug Administration recently made labeling changes restricting use of anti-EGFR monoclonal antibodies in colorectal cancer to patients who test negative for KRAS mutations.

Overlaying the importance of determining KRAS status is that labs using standard polymerase chain reaction (PCR) techniques face challenges in detecting mutations in tumors with a low percentage of mutant cells in a background of non-malignant cells. Consequently, tumor enrichment strategies are “critical,” according to Zhuang Zuo, MD, PhD, assistant professor of hematopathology at the University of Texas M.D. Anderson Cancer Center in Houston. “The samples we receive in the lab are a mixture of cancerous and normal cells, and this enrichment will increase the chance that we’ll be able to detect a mutation even though it’s in a very small percentage of the sample we’re reviewing.”

One promising amplification technique is fast co-amplification at lower denaturation temperature-PCR (COLD-PCR). Researchers at the University of Washington in Seattle recently reported success in implementing this method (BMC Clinical Pathology 2010, doi:10.1186/1472-6890-10-6). Because it runs at a lower denaturation temperature than standard PCR, fast COLD PCR favors amplification of mutant over wild-type alleles, according to lead author, Colin Pritchard, MD, PhD, chief resident and molecular diagnostics fellow in the department of laboratory medicine. “It’s probably just a slight favoring of more denaturation of mutant template than wild-type on each successive round of PCR, but because PCR is exponential, COLD-PCR really takes advantage of that—it’s like compounding interest,” he explained. “Even if you get just a few percent, over multiple rounds of PCR it becomes a huge enrichment of mutant over wild-type signal. We’ve observed up to a 100-fold mutation enrichment using COLD-PCR compared to regular PCR, but more typically for KRAS mutations we see about a 10-fold enrichment.”

Both Pritchard and Zuo, who was not involved in the study, became interested in COLD-PCR following the first report of this technique in 2008. Since then, at least 15 published studies have used COLD-PCR, and a growing number of labs are trying the method, according to Zuo. Fast COLD-PCR, which favors G:C to A:T mutations, is less complicated and takes only 1–2 hours of instrument time compared with full COLD-PCR, which theoretically enhances detection of any type of mutation, but takes 5–8 hours.

Pritchard and his colleagues tried the technique to investigate how well it performed in comparison to both melting curve analysis without COLD-PCR and traditional Sanger sequencing. They analyzed 61 formalin-fixed paraffin-embedded colorectal cancer tumor specimens. Of these, all three techniques detected KRAS mutations in 29 of 61 specimens and wild-type in 28. However, fast COLD-PCR detected mutations in an additional four samples that neither traditional melting curve analysis nor Sanger sequencing detected.

“We felt like we had good tumor enrichment in all the samples already because we performed dissection of the tumor area prior to preparing DNA. The regular melting curve analysis had a sensitivity of about 10 percent mutant allele on a wild-type background, which we thought would be sufficient. It’s better than direct Sanger sequencing, which is about 20 percent,” explained Pritchard. “However, to our surprise, we found that with fast COLD-PCR these four specimens were reclassified as positive. It wasn’t subtle; it was obvious that they were positive with COLD-PCR, and they were clearly negative when we used regular PCR.”

The findings underscore the importance of adopting robust tumor enrichment strategies in cancer mutation analysis, according to Pritchard. “I’d like to emphasize that tumor enrichment by microdissection is not enough,” he said. “There may be a conception out there among labs that yes, I know my method has a detection sensitivity of only 10 percent, but I’m making up for that because I’m carefully evaluating the tissue I’m selecting to do my testing on. We believed that and were proven wrong.”

Pritchard also stressed the clinical consequences of using enrichment methods with inadequate sensitivity. “Our studies and several others have shown that when you use less sensitive methods you miss about 10 percent of the positive cases. Clinically, that’s quite significant because mutation analysis often is being used as the sole determinant of targeted therapy decision-making in colorectal cancer patients,” he explained. “Physicians are making treatment decisions with this test, which has implications for the survival of patients and costs to the healthcare system. That’s why KRAS mutational analysis can be considered a high-risk test.”

Zuo concurred that mutation detection analysis has taken on new importance, making accurate methods all the more essential. “At our institution, there’s been an exponential increase in the use of these tests. Every colon cancer patient we see gets screened for KRAS status, and we also use KRAS status to monitor for early relapse, so patients are tested not just once, but multiple times,” he indicated.

While Zuo and Pritchard share enthusiasm for fast COLD-PCR, Zuo cautioned that labs could face some challenges in implementing the method. “Theoretically, it’s supposed to be very straightforward to develop and you don’t need to invest in new equipment,” he said. “However, in reality it’s not that straightforward. It can be tricky to find the optimized critical denaturation temperature. As for now, there’s no software or mathematical way to predict which temperature will work best for you, so it can be frustrating and time-consuming to find the best denaturing temperature.”

Pritchard explained that he too expected fast COLD-PCR would be difficult to implement, but found that not to be the case. “I had read about the method, and thought, this sounds great, but it probably took years for them to work out all the conditions. However, the first day I tried I found the critical denaturation temperature,” he said. Pritchard and his colleagues included in their paper tables of primer and probe sequences, as well as the fast COLD-PCR cycling program. “We thought that if people wanted to use our method, they’d need all the details of the acquisition modes, ramp speeds, and the like that are typically not published, even in methods papers. We wanted to make it as easy as possible for other labs to incorporate this,” he said.

Zuo stressed that the authors’ procedures and critical denaturation temperature are specific to the Roche LightCycler instrument. For example, Zuo’s lab uses a different instrument and has found 80ºC to be the optimal temperature, whereas Pritchard and his colleagues used 81ºC. Setting aside any challenges involved in identifying the critical denaturation temperature, Zuo strongly endorsed both Pritchard’s specific method and the overall fast COLD-PCR technique. “I encourage everyone to give it a try. This technology’s really nice, and I think revolutionary,” he said. “If it’s done right, it can really enhance detection of mutations.”

Join Us On LinkedIn

 Join us for a discussion of this article, and interact with your peers, on LinkedIn.

Rate this page:       
Page Access: