June 2017 ILSH: Hawaii - Hot Topics

The ISLH meeting took place this May in beautiful Hawaii. Not only was the weather hot, but the topics were also. Here are a few of the highlights!

Optimizing Approaches to molecular testing for bleeding disorders

In the majority of cases: diagnosis and clinical management of inherited bleeding disorders are done via laboratory tests - can precisely an inexpensively diagnosis bleeding disorders. Use molecular testing for confirmation of uncertain phenotype diagnosis; eg. In mild quantitative deficiencies, can perform molecular testing in particular if results are equivocal. When 300 samples were looked at using ThromboGenomics 63 gene platform, it validated 100% of 159 cases, and in 61 suspected phenotypes 92% of variants were identified. The diagnosis of inherited bleeding disorders will rely on standard coagulation test performance, however molecular testing will provide supplementary information to confirm diagnosis.

Controversies in Diagnosis in TYPE 1 vWD

Von Willebrand factor is synthesized in endothelia cells and megakaryocytes, stored in Weibel Palade bodies. Type 1 is a mild to moderate quantitative disorder found in 80% of patients.

Why is so important to get this right? Patients are symptomatic with a prevalence in 1 in 100 meaning about 34, 000 Canadians are affected however about 4000 have been diagnosed. Symptoms are treatable. On the other hand if you diagnosis someone who is unaffected, they are labeled with a genetic disease and exposed to unnecessary and inappropriate testing.

Are Bleeding Assessment Tools (BAT) useful in diagnosing vWD? Started with VINCENZA BAT in 2005 - 17 pages long - takes about 40 minutes to administer - others have been developed- ISTH - BAT- or even one that is self administered. If they have a low BAT that is a good NPV that patient doesn't have VWD.

Should genetic testing be done routinely? TYPE 1 vWD - looked at the spectrum of genetics and there is no one type that is prevalent for the mutation, therefore have to sequence the entire gene. Also need to consider the Influence of blood type in vWF levels in values of > 30% in particular in blood group O patients. This can be a contributing factor, possibly vWF may be cleared quicker. Also, evaluating inheritance patters within a family is challenging.

What cut off of VWD be used? Diagnosis of vWF level < 30% are classified as having vWD. While patients with levels between 30-50% called low vWF- (Based on) unpublished data shows individuals with low vWF bleed, have a positive bleeding scores. Within that group, vWF levels of 30-40% versus 40-50% don't have difference in bleeding scores. The lower, vWF the higher BAT scores.

Additionally, vWf normalizes over time since it increases with age. In those patient that had low values in the past up to 60% corrected into the normal range. What do you do with this? Does the bleeding phenotype go away? Probably the phenotype doesn't go away, needs to be evaluated. Also, should we use the category of LOW vWD? Not labelled as a disease- therefore may present a barrier to care- and patients may NOT get treated. Lots of questions that just add to the controversies in diagnosis.

D dimer

D dimer testing is complicated, reagents have different monoclonal antibodies, there are different reporting ranges, different standards used on different analyzers. The test has a sensitivity of between 69-97% with a specificity of between 43-99%. Testing results in a large number of positive ddimers without the diagnosis of VTE. When patients are scored and have a low test preprobability one can exclude VTE, however in patients that have a moderate to high risk, you can't rely on the d-dimer. When using the d-dimer threshold set by the manufacturer, it is set with the knowledge in mind that keeping it on the lower side will increase the sensitivity, however this decreases the specificity.

When looking at studies on patients with suspected DVT, the same threshold is used regardless of age. It is known that d-dimer increases with age. Using an age adjusted threshold can improve the specificity. The formula suggested is to take the age x 10 = threshold. This means a 65 year old would have a threshold of 650 (versus the standard threshold of 500). This could increase the specificity of the test. It was also suggested to reduce the number of ddimer tests used by not performing the assay on patients with a PTP of moderate and high.

Current Issues and Challenges in the laboratory diagnosis of hemophilia:

Does assay methodology matter? There can be a 2 fold difference in results in discrepant non-severe hemophilia. The one stage (OSA) or Chromogenic assay (CSA) may be higher and has been reported in up to 30% of individuals in non hemophila A, and also B. This is consistent between family members and all individuals bearing the same mutation- therefore has a molecular basis and has an impact on the different activity assays.

Discrepancy is due to missense mutation which may lead to missed diagnosis or misclassification of hemophilia. 11% of these patients have been reported as mild hemophilia, may have a significant bleeding tendency- low value correlates better with the bleeding tendency. The aPTT may not be a reliable screening test based on the sensitivity of the reagent. If the reagent has a poor sensitivity (<30%), a prolonged result may only happen when activity is very low reflecting factor VIII levels below 25% while the IX < 15% before you can screen for to hemophilia. Initial testing should include BOTH OSA and CSA.

Pitfalls associated with monitoring FVIII and IX replacement therapy:

Several new factor replacement products are new to the market and not all tests are optimal for recovery. In several rFVIII products, such as BAX 855 the field study demonstrated that there were no aPTT reagent dependent recoveries observed. While in studies that looked at CSL627 it was found that the OSA was almost 50% lower than the CSA regardless of the aPTT reagent used. It was suggested to use a correction factor of 2 if using the OSA. The CSA reflected correct levels.

Bay 94-9027- Bayer in house study- silica based reagents significantly underestimated FVIII levels. Ellagic acid reagents recovered well, as did CSA testing. Additional studies are pending.

When looking at FIX products: rFIX-FC, rIX-FP, N9-GP- some non linearity was observed in the OSA when using a PNP calibrator. Also underestimation was seen in kaolin aPTT. The package insert states that OSA kaolin based reagents underestimate FIX, therefore kaolin based and low sensitive LA reagents should be avoided.

The data shows many of the new agents are reagent depend and recovery of appropriate levels can be problematic. However the CSA is a viable option. A problem with this assay is the limited availability of IVD approved CSA in the US- 3 assays are IVD approved, only 1 kit evaluated by FDA for use on automated coagulation analyzer. FIX are only labeled as research use only with limited instrument application.

Laboratories need to be aware of these new products and the issues that have been seen with the OSA due to reagents. It has been demonstrated that selected rFVIII and IX are reagent dependent as seen in the OSA, versus the CSA which appears to be more consistent.

So here is a sampling of some of the wonderful lectures that happened at ISLH!