What to do about unclear Her2 test results

orange1

Often women post on this forum that their Her2 test results were equivocal (not clearly positive or negative.) Below are two articles that address this issue. The first article explains that regardless of degree of positivity of Her2+ testing, as long as the result is even slightly positive, women reduce their risk of recurrence significantly with Herceptin.The second article explains that even amoung expert pathologists, there are disagreements about how to interpret Her2 results - this includes not only IHC results, but also FISH results.Bottom line - if you test even slightly positive by any method (IHC or FISH), Herceptin is likely to signicantly decrease your chance of recurrence. If you test "negative" but are IHC 1+, do not accept this negative, get a FISH test.  In this study - all classes of tumors that showed evidence of Her positivity (IHC 1+, 2+ 3+ and FISH+) benefited significantly from Herceptin, including tumors that scored IHC 1+ or 2+ and did not subsequently test positive by FISH testing. From Journal of Clinical Oncology, 2007 ASCO Annual Meeting Proceedings (Post Meeting Edition).  Vol 25, No 18S (June 20 Supplement), 2007: 511AbstractBenefit from adjuvant trastuzumab may not be confined to patients with IHC 3+ and/or FISH-positive tumors:  Central testing results from NSABP B-31S. Pal, C. Kim...others511Background: Trastuzumab is a humanized monoclonal antibody targeted to HER2 protein and currently indicated for Her2 positive breast cancer defined by overexpression of Her2 protein (3+ IHC staining by HercepTest) or Her2 gene amplification (Her2/CEP17 ratio over 2 by PathVysion FISH assay).  These criteria were determined for advanced disease but have not been formally tested in the adjuvant setting.  We examined these tests' ability to predict benefit from adjuvant trastuzumab in NSABP B-31.  METHODS:  All available tumor tissue blocks from the B-31 trial were sujected to HercepTest and PathVysion assay as defined in the B-31 protocol.  Formal statistical test of interaction between HER2 levels measured by these two tests and benefit from trastuzumab was performed.  RESULTS: 207 or 1795 cases (11.5%) showed gene amplificiation as as determined by PathVysion, and 255 of 1662 (15.3%) showed overepression as determined by HercepTest.  161 of 1662 (9.7%) had neither gene amplificiation nor overexpression  There was a consistent benefit from trastuzumab in every subset defined by IHC or FISH.  No statistical interaction was found between DFS benefit from trastuzumab and level of protein (p = 0.26) or Her 2 gene copy number (p=  0.60).  Benefit was observed in patients with tumors that were negative for FISH and had less than 3+ staining intensity on IHC by HercepTest (RR= 0.36, p = 0.32).  CONCLUSION:  Current definition of HER2 overexpression/gene amplificiation based on data from adjvanced disease may need to modified for the adjuvant setting.In all categories of Her2+, regardless of degree or how measured, relative risk with Herceptin was less than half of what it was without.  And all results were highly statistically significant, meaning results were likely not due to random chance. From Medscape:HER2 Testing: Accuracy Remains an IssueKathy D. Miller, MDAuthors and DisclosuresPosted: 01/24/2011Happy New Year, everybody. It's Dr. Kathy Miller from Indiana University back with my first Medscape blog of the New Year. Today I have to take us back to a subject that many of us keep wondering when we will be able to stop talking about, and that is the importance of accurate HER2 testing. Will we ever get to a day when we can just get a single test result, have confidence in that result, and make treatment decisions on the basis of that result without constant retesting and second-guessing?The bad news is that we are a long way away from that, and the most recent evidence of how complicated this gets comes from a round-robin assessment of the pathology samples from 3 clinical trials: the North Central Cancer Treatment Group (NCCTG) adjuvant trial, a Breast Cancer International Research Group (BCIRG) adjuvant trial, and then an earlier BCIRG trial that also had samples available for testing. These samples were tested by both immunohistochemistry (IHC) and by fluorescence in situ hybridization (FISH). Slides were then sent in a round-robin fashion to 3 blinded pathologists. They had at least 2 or 3 interpretations of those samples on almost all of the 375 of 380 or so samples that they had available. Even with these expert pathologists reading the same slides (this gets around all of the issues of testing in individual laboratories and doing IHC correctly), this analysis only looked at how reproducible the pathologists' eyes were and their interpretations.So with these expert pathologists and IHC round-robin, 8% of the time there were discordant results between what they considered positive and negative. Now in this round-robin session, in this study if there was discordance, the pathologists then came together -- same place, same time -- and looked down through the same microscope tube at the slide at the same time and battled it out to come to a consensus. Here is what I find most fascinating, even with that discussion and looking at it at the same time: Just under half the time they still couldn't come to an agreement. The best you could get was a majority opinion of 2 out of 3 pathologists.Some pathologists and many medical oncologists in the community have thought that the way to get around the inherent subjectivity of IHC is to just do FISH testing on everything. It's all quantitated; you count the number of red dots, and the number of green dots; you do the math; you get the results. FISH testing was included in this study as well. A similar number of cases were sent for the round-robin review with FISH testing; 8% of the time there was no agreement (discordant results) between the pathologists. They couldn't all get together in the same way for review of those results; they were able to get together for review of about a third of them, and in these cases they got the same results. The remaining cases were restained and sent around for a second review, and still in roughly half of the cases, the pathologists could not come to a consensus.Testing remains an issue, and it is an absolutely crucial one for medical oncologists to understand. We make important decisions, potentially life-altering decisions about our patients' therapy on the basis of these results. We need to talk to our pathologists. We need to know who is doing the test; how they are trying; whether they are using standard procedures; who is interpreting the test; and if there is any question, the samples ought to be retested or the slides ought to be viewed by another pathologist. We should not be surprised if there is discordance, because 8% of the time, even with expert pathologists, they disagree. In some cases, even battling it out at the same time, they still disagreed. Any time human beings are required to look at something with their eyes and make interpretations, there will be differences; that is simply a fact of life. But in this case we can't afford not to get it right.Talk to your pathologists; retest the samples; be vigilant about requesting retesting; don't let the pathologist tell you that they always do it right because nobody always does it right, and this one is just too important.

lago

At my treatment center they did a FISH on my biopsy sample. They repeated the FISH test on my tumor after my BMX. In my case the results were the same both times (strongly postive). They never did the IHC test from what I could see.

septembersong

Orange1,

Thanks so much for posting this.

I can remember two things about this part of my own diagnosis: it seemed to take forever (actually about three weeks) to get the results of both tests; and the doctors (at two different treatment centers) did a very poor job of explaining both the issues and the processes. With so much hanging in the balance, doctors need to do a better job of educating stressed and confused patients who need to make critical decisions about treatment. These articles give some insight into why this doesn't happen more often.

Thanks for sharing.

Ann