Response to Second Letter on Response Letter Regarding “Plasma Concentration of Thrombopoietin in Dogs With Immune Thrombocytopenia”

Abstract

We appreciate Dr. Wun's ongoing interest in our publication and hope he feels free to email or phone the authors directly if this response leaves any remaining questions.

As we describe in section 3.3, the addition of cryopoor plasma to the assay buffer was a means to correct the non-quantitative recovery of recombinant canine TPO (rcTPO) due to high background plasma absorbance values. When varying amounts of rcTPO were added to canine plasma samples and the OD values for those samples were compared to a standard curve constructed using rcTPO in non-proteinaceous assay buffer, the measured concentrations did not match the expected concentrations. Once the matrix effects associated with the presence of proteins other than TPO in canine plasma were accounted for by adding a small quantity (1:4 ratio) of canine cryopoor plasma to the assay buffer, the recovered concentrations of rcTPO matched those expected as described in section 3.5.

Dr. Wun proposes that this use of cryopoor plasma within the diluent for standard curve samples might impair the detection of endogenous TPO in ITP patient samples. We have previously acknowledged that the assay lacks precision at very low TPO concentrations. However, the relevant point is that dogs with thrombocytopenia should have high plasma TPO concentrations and hence the assay's stated limitation is less of concern when the assay is applied to the primary target population. In our publication, all five dogs with induced ITP had dramatic increases over time in endogenous TPO concentrations compared with their low baseline values, as expected from studies in rodent models. This observation indicates that the assay is fit for purpose to investigate mechanistic causes of thrombocytopenia in dogs.

Separately, we believe Dr. Wun may have misunderstood some aspects of the TPO assay results presented in our paper and their clinical applications. The speculation put forth that an interferent unique to ITP samples could prevent quantitative recovery of rcTPO when spiked into ITP samples is irrelevant, because we measured endogenous TPO concentrations in dogs with ITP. Similarly, we did not suggest that the assay could be used to differentiate recombinant from endogenous TPO nor is therapeutic monitoring of rcTPO an actual or potential clinical use of the assay. Although it is theoretically possible that plasma from a given ITP dog might contain a substance that could interfere with detection of endogenous TPO, it is highly unlikely that plasma from every dog with ITP would contain such a substance, and it is unclear which interfering substance Dr. Wun speculates to be present. It is also neither practical nor standard to test every sick patient sample with spike and recovery of standard to ensure the absence of a theoretical interferent.

The rationale for the use of TPO receptor agonists in humans with ITP was based on the observation that ITP patients had similar plasma TPO concentrations as people with normal platelet counts. Our study found that endogenous plasma TPO concentrations of 56 of 58 dogs with naturally occurring ITP overlapped with those of dogs with normal platelet counts. This observation provides the same rationale for exploring use of these agonists in dogs. We look forward to treatment trials that can define a role for TPO receptor agonists in managing ITP in dogs.

The authors declare no conflicts of interest.