Multi-Analyte Thyroid Function Testing by LC-MS/MS: A new way to look at thyroid function?

Thyroid hormones play an essential role in metabolism, growth and development.
Thyroid hormone metabolism was considered to be important for recycling of iodine
and clearance of thyroid hormone metabolites, but the thyroid hormone metabolites
itself were regarded to be inactive. Nowadays, thyroid hormone metabolites downstream
from the two thyroid hormones T3 and T4 are being suggested to have several
physiological roles. To gain insight into the clinical effects of the metabolites and to
determine whether thyroid hormone metabolites give us meaningful information to
better understand pathophysiological processes, we developed a multi-analyte assay
using mass spectrometry. The metabolites that were combined in the multi-analyte assay
was determined by the clinical question we wanted to answer and the structural and
chemical properties of the metabolites.
The research which is presented in this thesis focuses on the method
development, including analytical and clinical validation, to quantify total
concentrations of two thyroid hormones and seven thyroid hormone metabolites in a
single run using liquid chromatography-tandem mass spectrometry (LC-MS/MS). The
method was validated according to CLSI and EMA guidelines. The clinical validation
was performed in a cohort of healthy individuals, patients with differentiated thyroid
cancer and patients with acute coronary syndrome.
Chapter 1 provides a general introduction on thyroid hormone metabolism, the
physiological roles attributed to the studied thyroid hormone metabolites and the
different analytical techniques used to measure thyroid hormones in healthy individuals.
Chapter 2 focuses on sample preparation strategies for quantification of thyroid
hormones and thyroid hormone metabolites by targeted LC-MS/MS and describe a new
mass spectrometric method for the quantification of seven iodothyronines and two
iodothyroacetic acids in human serum.
Chapter 2 Part A highlights sample preparation strategies, since method
development of a multi-analyte assay is challenging due to diverse chemical properties of
the metabolites and broad concentration ranges (picomolar to millimolar) among
metabolites. We discuss protein precipitation, liquid-liquid extraction and solid phase
extraction and also focus on other key factors to consider, such as adsorptive loss to
standard laboratory ware, availability of analyte-free biological matrices, internal
standards, reconstitution and stability during sample preparation. We advise to take four
considerations into account, when developing a multi-analyte assay for thyroid hormone
metabolites. First, use glass vials and glass inserts for single-use due to adsorptive loss to
polypropylene and to minimize cross-contamination. Second, use an extensive and
selective sample preparation strategy e.g. combine a protein precipitation step with a
solid phase extraction step with a mixed-mode sorbent to increase selectivity and obtain
a clean sample. Third, check the internal standard for impurities of the non-labeled
compound and of other non-labeled and labeled compounds that are quantified in the
metabolite panel. Fourth, determine the needed equilibration time of the internal
standard to ensure proper correction of losses during the entire sample preparation.
In Chapter 2 Part B, the method development is described of a mass
spectrometry-based assay to quantify seven iodothyronines and two iodothyroacetic
acids in human serum in a single run. Subsequently, reference values were determined in
healthy individuals. Extensive sample preparation and the use of 13C6-labelled internal
standards for eight out of nine thyroid hormones and its metabolites was very important
for a sensitive and robust analytical method. This LC-MS/MS method is unique as it
contains three additional thyroid hormone metabolites compared to previously
published thyroid hormone metabolite panels and combines two metabolic pathways,
iodothyronines and iodothyroacetic acids. We found lower concentrations for 3,5-T2
and 3,3'-T2 in healthy individuals with our method compared to previously reported
concentrations measured with LC-MS/MS. In our reference group, TA3 could not be
detected and a large variation in TA4 concentrations was found. A large TA4 variation
has also been reported in other euthyroid patient cohorts.
In Chapter 3, we focused on the binding characteristics of thyroid hormone
metabolites with thyroid hormone distributor proteins. In addition, we determined the
association of thyroid hormone distributor proteins with thyroid hormones and thyroid
hormone metabolites using in vivo concentrations. We found that the distribution of
thyroid hormone metabolites between distributor proteins differs from that of T4 and
T3, which predominantly bind to TBG. The predominant distributor protein of 3,3'-
T2 and rT3 is albumin, of TA3 is TTR and albumin and of TA4 is TTR. For TBG, the
rank order of affinity was T4>TA4=rT3>T3>TA3=3,3'-T2>3-T1=3,5-T2>T0 (IC50-
range: 0.36 nM to > 100 μM) and for TTR the rank order of affinity was
TA4>T4=TA3>rT3>T3>3,3'-T2>3-T1>3,5-T2>T0 (IC50-range: 0.94 nM to >100
μM). A positive association was found for TBG with T3 and T4, but not for TTR or
albumin with T3 and T4. TBG, TTR and albumin were not associated with T0, 3-T1,
3,3'-T2, rT3 and TA4. Therefore, serum TBG, TTR and albumin concentrations
within the reference interval do not influence serum concentrations of T0, 3-T1,
3,3'-T2, rT3 and TA4.
The clinical validation of the method is described in Chapters 4 to 6.
In Chapter 4, we investigated thyroid hormone metabolite concentrations
across different thyroid states in a cohort of patients treated for differentiated thyroid
cancer. We used this patient cohort to answer two questions; 1) how does thyroidectomy
affect thyroid hormone metabolite concentrations? and 2) does LT4 supplementation
therapy restore thyroid hormone metabolite concentrations in patients without a
thyroid gland? In our study, we found that thyroidectomy causes a decrease in all thyroid
hormone and thyroid hormone metabolite concentrations. Additionally, we found that
LT4 supplementation therapy restores thyroid hormone metabolite concentrations
following the same trend as T4. The latter suggests that thyroid hormone metabolites in
general are predominantly formed via peripheral extrathyroidal metabolism. In patients
without a thyroid gland, T3 production is also solely dependent on peripheral
extrathyroidal metabolism. Consequently, more LT4 should be supplemented until
subclinical hyperthyroid levels are reached to ensure that T3 is within the reference
interval.
In Chapter 5, we studied the relationship (cross-sectionally and longitudinally)
between thyroid hormones and its metabolites and quality of life across different thyroid
states. In addition, we sought to identify a specific thyroid hormone metabolite that can
explain the persistent symptoms in patients with hypothyroidism on LT4 replacement
therapy. With the cross-sectional analysis, we did not find a relation between thyroid
hormones and thyroid hormone metabolites with Quality of Life at any of the visits.
With the longitudinal analysis, in general higher concentrations of T0, 3-T1, 3,3'-T2,
T3, rT3 and TA4 were associated with significantly less complaints in specific Quality of
Life domains compared to those with lower concentrations of these metabolites.
However, we did not find any metabolite responsible for the persistent complaints in
hypothyroid patients on LT4 replacement therapy.
In Chapter 6, we determined whether thyroid hormone metabolite
concentrations and thyroid hormone metabolite ratios are associated with an increased
risk of major adverse cardiovascular events and mortality in patients admitted for acute
coronary syndrome. The results show that higher T3/T4, T3/rT3 and T4/rT3 ratios are
associated with a lower risk of major adverse cardiovascular events and mortality in the
unadjusted model. When adjusting for age and sex the associations of T4/rT3 ratio with
major adverse cardiovascular events and T3/T4 ratio and T4/rT3 ratio with mortality
lost significance. A higher rT3/3,3'-T2 ratio is associated with a higher risk of major
adverse cardiovascular events and mortality, but lost statistical significance when
adjusting for age and sex. In non-thyroidal illness, D1 is downregulated and D3 is
upregulated. Our results on thyroid hormone metabolite ratios indicate that the thyroid
hormone metabolite pattern in patients admitted for acute coronary syndrome
resembles non-thyroidal illness. In addition, higher T0 concentrations are associated
with an increased risk of major adverse cardiovascular events and mortality in these
patients.
Finally in Chapter 7, we discuss the results of the studies presented in this thesis,
together with the possible implications of these studies, the limitations of the studies as
well as future perspectives.