Chronic myeloid leukaemia (CML) is a clonal myeloproliferative neoplasm marked by uncontrolled myeloid cell growth in the bone marrow. Tyrosine kinase inhibitors (TKI) are the primary treatment for CML but can cause side effects, including thyroid function test abnormalities. AIM: This study aimed to evaluate the thyroid function abnormality produced by the drug TKI in patients of CML and whether the thyroid function abnormality has a potential role to serve as a biomarker for clinical tumour response. Methods: Diagnosed patients of CML receiving TKI and who were in remission, i.e. patients in molecular response, were taken as the study group and were subjected to thyroid function tests. The thyroid function test results were observed and were associated with the molecular response pattern. Results: The study group was about 100 patients diagnosed with CML on TKI in molecular response, both major and deep molecular response. In the study group, 61/100 patients were in deep molecular response (DMR), and 39/100 patients were in major molecular response (MMR), with 65/100 patients receiving imatinib and 35/100 patients receiving nilotinib. The thyroid function test done on the study group showed that 75/100 patients had thyroid function abnormality, with hypothyroidism (70/100) being the most common abnormality. A statistically significant association was found between S.TSH levels (p<0.001) and S. free T3 level (p=0.047) with the molecular responses. also, a statistically significant association was found between hyperthyroidism (p=0.008) and a normal thyroid function test (p=0.013) with the molecular responses [p value <0.05 was considered statistically significant]. Conclusion: Our study confirms that tyrosine kinase inhibitors (TKIs) affect thyroid function, particularly in patients with chronic myeloid leukemia (CML). Hypothyroidism is one of the most common abnormalities observed.
Tyrosine kinase inhibitors (TKIs) are a group of drugs that disrupt signal transduction pathways of protein kinases through various modes of inhibition. The approval of the first TKI, imatinib, in 2001 marked a significant advancement in the treatment of chronic myeloid leukemia (CML), leading to the evolution of cancer chemotherapy into the current era of targeted therapy. TKIs are now considered the standard of care for several cancers, with CML being a prominent example where TKIs are the preferred treatment regardless of the stage of the disease.
TKIs function by inhibiting tyrosine kinase enzymes, which can be classified into receptor tyrosine kinases, non-receptor tyrosine kinases, and dual specificity kinases. Currently, the United States Food and Drug Administration (FDA) has approved over 50 FDA TKIs for use. Adverse drug events of TKIs are generally dose-dependent and are associated with specific side effect profiles unique to each drug. Due to similarities in drug targets, different classes of TKIs may produce similar side effects. Common side effects shared by TKIs include cutaneous drug reactions, fatigue, fever, gastrointestinal disturbances, cardiovascular side effects (such as hypertension), and endocrine-related issues like thyroid dysfunction.
CML is a myeloproliferative neoplasm caused by the reciprocal translocation between chromosomes 9 and 22, resulting in the formation of the BCR-ABL1 fusion gene, which drives the disease process. CML represents a rare haematological malignancy with remarkable responses to targeted therapy. TKIs are presently used in the management of CML patients at any stage. There are 6 TKIs approved and in use for CML treatment, categorized into three generations:
1st generation – Imatinib
2nd generation – Dasatinib, Nilotinib, Bosutinib, and Radotinib
3rd generation – Ponatinib
These drugs vary in their pharmacological profiles. Clinical, haematological, cytogenetic, and molecular responses are used to assess the response of CML patients to these drugs and reflect the leukemic burden in these patients. TKIS are known to produce thripid dysfunction, and it is a well-documented side effect of TKIS used to treat solid tumours, but there is a paucity of studies to validate this side effect in haematological malignancies. This study tried to confirm the hypothesis that thyroid dysfunction is induced by tyrosine kinase inhibitors used to treat CML, and whether there is any correlation with the molecular response attained by these patients.
Study design:
The study was a cross-sectional observational study that followed a protocol approved by the institute's ethics committee before its execution.
The flow of the study:
The study was conducted in a tertiary care hospital in New Delhi, India. Patients were follow-up cases of chronic myeloid leukaemia on tyrosine kinase inhibitors in remission. The records of these patients were reviewed, which especially included whether the patient was on any anti-thyroid drug or thyroid supplementation drug, and the type of molecular response the patient was in. We include patients of the age of more than 18 years of any gender with chronic myeloid leukaemia on tyrosine kinase inhibitors for more than 18 months with the molecular response (major and deep molecular response). we exclude Patients with pre-existing thyroid alterations based on previous records, Patients on thyroid hormone supplement therapy or anti-thyroid drugs or any other drug-altering thyroid function test other than TKI such as amiodarone, lithium, selective serotonin reuptake inhibitors, rifampin…. etc., Patients with nephropathy of any cause, chronic liver illness was also excluded from the study. Eligible patients were included in the study and were subjected to a history targeted at symptoms of hypothyroidism or hyperthyroidism with examination of the thyroid gland and thyroid function tests using chemiluminescent microparticle immunoassay. The thyroid function test included serum-free T3, free T4 and TSH levels. Along with the thyroid function test, molecular response, which was previously documented, was confirmed by running RT-PCR of the peripheral blood for BCR-ABL transcripts.
Interpretation of test results:
Hypothyroidism: elevated serum TSH level with reduced serum-free T4 level.
Hyperthyroidism: reduced serum TSH level with elevated serum-free T4 level.
Sub-clinical hypothyroidism: elevated serum TSH level with normal serum T4 level.
Molecular response (MR) was defined according to the European Leukemia Network guidelines – Major Molecular Response (MMR) was defined as BCR-ABL1 IS transcripts less than 0.1%, and Deep Molecular Response (DMR) was defined as BCR-ABL1 IS transcripts less than 0.01%. Both MMR and DMR patients were assessed for thyroid function.
The test results were interpreted and then subjected to statistical analysis.
Statistical analysis:
Sample size: The sample size was calculated using Openepi version 3.2 with a confidence level of 95%. The p-value was set at 34.8% from the study by Rodia R et al28. The absolute precision was defined as 10% (0.1). Z = 1.96; q = 1-p = 0.652.
n = Z2 pq / d2
The calculated sample size (n) is 87.
Analysis of data:
The categorical variables were presented as numbers, while the quantitative data were shown as means ± standard deviation (SD) and as medians with the 25th and 75th percentiles (interquartile range). The normality of the data was assessed using the Kolmogorov-Smirnov test. For cases where the data did not meet the criteria for normal distribution, nonparametric tests were employed. The following statistical tests were utilised for the analysis of the results:
Data entry was performed using Microsoft Excel 2021, and the final analysis was conducted with the Statistical Package for the Social Sciences (SPSS) software, version 29.0, produced by IBM, Chicago, USA. A p-value of less than 0.05 was regarded as statistically significant.
In our study, most of the patients were male, and the mean age of the population was 50.4 years. The youngest patient was 26 years old, and the eldest was 78 years old. The majority of the patients received imatinib, and most of the patients were in deep molecular response. Abnormal thyroid function test was observed in 75/100 patients with hypothyroidism being the most common abnormality (Table 1).
Table 1 CML patients’ characteristics and thyroid profile observed during TKIs therapy
y
*Hypothyroidism includes both sub-clinical and clinical hypothyroidism
DMR – deep molecular response, MMR – major molecular response.
Out of 75 patients with abnormal thyroid function tests, 70 were observed to have hypothyroidism. Among these, 37 patients were clinically hypothyroid, with 27 in deep molecular response and 10 in major molecular response. 33 patients were sub-clinically hypothyroid, with 24 in deep molecular response and 9 in major molecular response. Additionally, 5 patients were hyperthyroid, all of whom were in major molecular response. Finally, 25 patients had a normal thyroid profile (Figure 1).
Figure 1
Out of 37 patients with clinical hypothyroidism, 7 did not show symptoms, while the remaining 30 did. Among the 33 patients with sub-clinical hypothyroidism, 4 displayed symptoms, and the other 29 were asymptomatic. Additionally, 5 patients were diagnosed with hyperthyroidism, with 4 showing symptoms and 1 being asymptomatic. Notably, 2 out of the 4 hypothyroid patients displayed thyroid eye signs (Figure 2).
Figure 2
CH- Clinical hypothyroidism; SH – Subclinical hypothyroidism; N – Normal; HYP – Hyperthyroidism;
The results showed a significant association between thyroid function tests and molecular response, specifically deep and major molecular responses. There was a notable correlation between serum TSH level (p < 0.001) and free T3 level (p = 0.047) with both deep and major molecular responses. In this study, p < 0.05 was considered statistically significant (Table 2).
Table 2 Association of thyroid function test with molecular response of the study participants
Thyroid function test [Mean (SD)] |
DMR |
MMR |
t-value# |
p-value |
S TSH |
7.2 (3.1) |
4.9 (3.3) |
3.51 |
<0.001* |
Freet T3 |
4.9 (2.8) |
3.9 (1.9) |
2.00 |
0.047* |
Free T4 |
13.9 (7.1) |
11.6 (6.2) |
1.78 |
0.079 |
#Student’s t-test
*p-value <0.05 statistically significant
In analysing the thyroid profile alongside the molecular response, a significant association was discovered between hyperthyroidism and both molecular responses, with a p-value of 0.008. Additionally, a significant association was observed between patients with a normal thyroid profile and both molecular responses, with a p-value of 0.013 (table 3).
Table 3
Inference |
MMR |
DMR |
p-value |
|
CH |
Yes |
10 |
27 |
0.059 |
No |
29 |
34 |
||
HYP |
Yes |
5 |
0 |
0.008** |
No |
34 |
61 |
||
N |
Yes |
15 |
10 |
0.013* |
No |
24 |
51 |
||
SH |
Yes |
9 |
24 |
0.092 |
No |
30 |
37 |
*p-value
<0.05 statistically significant.
**Fisher’s exact p-value <0.05 statistically significant.
CH- Clinical hypothyroidism; SH – Subclinical hypothyroidism; N – Normal; HYP – Hyperthyroidism.
Chronic myeloid leukaemia is a clonal myeloproliferative neoplasm characterised by an unregulated expansion of myeloid cells in the bone marrow. The driving force behind the disease process is the reciprocal translocation between chromosomes 9 and 22, leading to the formation of the oncoprotein BCR-ABL1, which produces a dysregulated tyrosine kinase. The discovery of this dysregulated tyrosine kinase led to the development of targeted cancer therapy. Imatinib, the first tyrosine kinase inhibitor developed for use in CML, was approved in the year 2001. Since then, various tyrosine kinase inhibitors have been developed for the management of CML as well as other malignancies and non-malignant conditions. Tyrosine kinase has various side effects, ranging from simple GI disturbances to secondary malignancies, but one of the side effects of interest is thyroid dysfunction produced by tyrosine kinase inhibitors. TKIs have anti-proliferative and anti-angiogenic properties. The exact incidence of thyroid dysfunction caused by tyrosine kinase inhibitors (TKIs) used in the treatment of chronic myeloid leukemia (CML) is currently unknown. However, case series have indicated that up to 90% of patients receiving TKIs may develop thyroid dysfunction. In contrast, TKIs used for treating solid tumors show a very high incidence of thyroid dysfunction as well. The proposed mechanisms for thyroid dysfunction induced by tyrosine kinase inhibitors include thyroiditis, autoimmunity, inhibition of vascular endothelial growth factor receptor (VEGFR), and blockade of iodine uptake. In our study, we studied the thyroid profile in patients with CML receiving tyrosine kinase inhibitors. Our study was a cross-sectional observational study in which 100 patients who were diagnosed with chronic myeloid leukaemia were taken as the study group. All these patients were receiving tyrosine kinase inhibitors and were either in major or deep molecular response. The molecular responses of these patients were assessed by performing Real-time polymerase chain reaction of the peripheral blood of these patients. Patients with pre-existing thyroid function abnormalities were ruled out by going through their previous records of thyroid function tests and asking them whether they were on thyroid hormone supplements.
The annual incidence of chronic myeloid leukaemia (CML) is reported to be 0.7-1.0 cases per 100,000 population based on data from several European CML registries. The average age at diagnosis of CML is between 57 and 60 years, and the male-to-female ratio ranges from 1.2 to 1.7. The prevalence of CML is not well established, but it is estimated to be 10-12 cases per 100,000 population. In India, the reported incidence of CML has varied from 0.8 to 2.2 cases per 100,000 population, but these figures are based on a registry that doesn't distinguish between acute and chronic forms of myeloid leukemia. However, this incidence is lower compared to the data from registries in the US, Europe, and Australia. This raises questions about the incidence of CML in India, especially given the country's high population density. In our study most of the patients were male and the mean age of the population was 50.4 years. Usually in Indian centers even though 2nd generation TKI’s are preferred than imatinib in both attaining remission and side effect profile, imatinib is still the preferred first line drug. Imatinib is dosed at 400mg per day. In our centre, usually, patients who are newly diagnosed with CML are started on imatinib, but in some centres in India, patients are directly started on second-generation TKI such as nilotinib, which is dosed at 300mg twice daily. In our study out of 100 patients 65 patients were on imatinib and 35 patients were on nilotinib. In all these patients, there was no prior history of use of any other TKI. Hence, from our study, we concluded that the majority of our patients were started on imatinib as first-line therapy, and a significant portion of patients were started on nilotinib as first-line drug therapy. We attribute this selection of TKI 1st gen VS 2nd gen to the drug affordability of our patients. Measuring the molecular response is the current modality of monitoring response to therapy when compared to earlier used methods such as monitoring spleen size, blasts in peripheral blood, etc. once the patient has been initiated on tyrosine kinase inhibitors then RT-PCR of peripheral blood is usually done at 3 months interval until the patient reaches the desirable molecular response. The ideal or desirable molecular response is the major molecular response achieved at the 12th month after initiation of therapy. After achieving such a molecular response, monitoring of the response is usually done at 2-3 times per year. Observations from our study showed majority of the patients in both deep and major molecular response were on imatinib. TKI causes endocrine side effects such as thyroid dysfunction, hyperparathyroidism, adrenal insufficiency, hypo/hyperglycemia, hypogonadism, alteration of bone mineral density, and the list keeps adding on. In our study, 75/100 patients had thyroid function test abnormality; of these patients, 37 were clinically hypothyroid, 33 were sub-clinically hypothyroid, and 5 were hyperthyroid. So, our study concludes that thyroid dysfunction is seen with the use of tyrosine kinase inhibitors, and hypothyroidism is the most frequently observed thyroid function test abnormality. In our study, 37 patients were clinically hypothyroid, of whom 30 patients had symptoms of hypothyroidism and 7 patients were asymptomatic for hypothyroidism. 33 patients were sub-clinically hypothyroid, with 4 patients having symptoms of hypothyroidism, and the rest of the 29 patients were asymptomatic. 5 patients were hyperthyroid, of which 4 were symptomatic and 1 was asymptomatic. Interestingly, in our study, it was observed that 2/4 patients with symptoms of hyperthyroidism had thyroid eye signs. These eye findings were then re-confirmed by specialist opinion from an ophthalmologist. Our study concludes that most of the patients with clinical hypothyroidism and hyperthyroidism are symptomatic for thyroid dysfunction. There is only one study in the literature that associated molecular response with drug adverse effects in CML patients; hence, our study is a standalone study. Our study concludes that there is a significant association between thyroid function test and molecular response with significant association between thyroid profile and molecular response.
Limitations:
The duration for which the patients were on tyrosine kinase inhibitors could not be traced; hence, a linear correlation between the thyroid dysfunction and the duration of therapy could not be drawn. The selection of patients into the study was made on the basis of reviewing previous records of thyroid function tests and whether the patients were on any anti-thyroid or thyroid supplementary drugs. Hence, there was no available baseline thyroid function test for comparison. The protocol was a cross-sectional observational study; hence, an etiological workup was not considered. The protocol was designed to only consider whether symptoms of hypothyroidism or hyperthyroidism were present. Individual symptoms could not be taken into account, as symptoms of thyroid dysfunction are not specific.