Treatment, Platelet Count Pattern and Comorbidities Among Primary Immune Thrombocytopenia (ITP) Patients Seen in Routine Clinical Practice: Data from the United Kingdom Immune Thrombocytopenia (UK ITP) Registry

Introduction

Primary ITP is an acquired bleeding disorder which is diagnosed by elimination of secondary causes of thrombocytopenia (<100 x 10⁹/L). In a platelet depleted state, patients experience or are at risk of bleeding. Treatment responses can vary widely and treatment-associated morbidity is a key factor in management in order to reduce complications and mortality, and improve quality of life.

Aim

The aim of this study is to describe treatment, platelet count pattern and prevalence of certain comorbid events of patients who were diagnosed after the introduction of the 'International consensus report on the investigation and management of primary immune thrombocytopenia', particularly focusing on those who did not require any treatment, those who received only 1^st line treatment and those who also required 2^nd line treatment.

Method

The UK Adult ITP Registry recruits adult participants (n=2700, to date) nationally through a network of around 80 centres. This study was limited to participants diagnosed from 2010 who were followed up until 31^st January 2015. Various statistical analysis were carried out, including meta analytical technique (using random effect modelling) for pooling multiple platelet counts [inc. 95% confidence interval (CI)] available per participant for pre- and post ITP periods in time.

Results

This study cohort consisted of 704 participants [female 55.3%; mean & median diagnosis age 51.9 & 54.1 years (IQR 33.7, 68.4); peak diagnosis age groups 18-29 years (18.5%) & 60-69 years (17.3%); 77.7% European ancestry].

One fifth (n=138) of participants did not receive any treatment and had pooled mean (PM) platelet counts of 58.7 (95% CI 46.3, 71.0) a week before and 67.2 (95% CI 60.7, 73.8) a week after ITP diagnosis. Forty one percent (n=292) had a 1^st line treatment (steroids, IVIg and/or anti-D) only whereas 35.5% needed to be given a 2^nd line treatment too (rituximab 50.2%, azathioprine 39%, mycophenolate 37.5%, romiplostim 32,7%, eltrombopag 21.1%, splenectomy 13.2%, others <5%). Distribution of 1^st line treatment for the former and latter sub cohorts were: prednisolone 88.0% vs. 92.8%, IVIg 28.8% vs. 65.7% and, anti-D 1.0% vs. 5.6%

The historical PM platelet counts (over 14 days before ITP diagnosis) were non-significantly higher for those who required only 1^st line treatment [131.1 (95% CI 114.6, 147.6)] when compared against those who required 1^st and 2^nd line treatment [104.6 (95% CI 85.7, 123.4)]. A week before diagnosis, PM platelet counts were not dissimilar 31.1 (95% CI 24.8, 37.4) vs. 31.8 (95% CI 16.4, 47.3)] but within the first week of treatment 59.6 (95% CI 45.1, 74.1) vs. [45.9 (95% CI 36.7, 55.1)], 1 month [138.1 (95% CI 120.6, 155.6 vs. 70.1 (95% CI 60.7, 79.4)], and up to 3 months after [157.0 (95% CI 142.9, 171.1) vs. 98.9 (95% 86.5, 111.3)] platelet count response seemed to be better during the 3 months for those who only received 1^st line compared against those who went on to receive 2^nd line treatment.

There were more females in the '1^st line treatment only 'cohort (58%, p<0.001). No gender differences were found among those who did not require any treatment or those who ended up having 2^nd line treatment. There was no dissimilarity in age at diagnosis or distribution of comorbidities among these sub cohorts.

Conclusion

This study was beneficial in identifying pertinent patterns for more in-depth analysis and interpretation. A third of participants were found to require a second line treatment at some point after their diagnoses. Since 2010, splenectomy still remained a treatment choice whereas romiplostim and eltrombopag are utilised by 20 to 30% of the cohort. Participants who ended up having a second line treatment tended to have a lower platelet count profile before and within 3 months of ITP diagnosis; this later period usually coincided with 1^st line therapeutic intervention. Whether this is predictive of their poor responses and requiring 2^nd line treatment will be analysed further while considering other pertinent characteristics. However, a larger cohort is needed to fully assess effects of different treatments on outcomes and morbidity, through wider collaboration.

Figure

View largeDownload slide

Figure

View largeDownload slide

Close modal