Cardiac morbidity and the cause of death in elderly patients with prostate cancer and incidental cardiac uptake on bone scintigraphy

Cardiac transthyretin amyloidosis (ATTR) is a possible incidental finding on bone scintigraphy imaged due to prostate cancer. We investigated its significance in 1426 elderly prostate cancer patients (>70 years) who underwent bone scintigraphy in three nuclear medicine departments in Finland. Patients with Perugini grade two or three uptakes were considered positive for cardiac uptake. Heart failure diagnoses and pacemaker implantations were collected from the hospital's records. Mortality data were gathered from the Finnish national statistical service (Statistics Finland). The Median follow‐up time was 4 years (interquartile range: 2−5 years). Cardiac uptake was detected in 37 individuals (2.6%), and it was associated with an elevated risk of both overall and cardiovascular death in univariable analysis. However, cardiac uptake did not predict overall mortality in the multivariable analysis when adjusted to age, bone metastases or the diagnosis of heart failure (p > 0.05). The risk of heart failure was higher in patients with cardiac uptake (47% vs. 15%, p < 0.001), while the risk of pacemaker implantations was not elevated (5% vs. 5%, p = 0.89). In conclusion, cardiac uptake on bone scintigraphy imaged due to prostate cancer is associated with an elevated risk of heart failure and both overall and cardiovascular death. However, cardiac uptake was not independently associated with overall mortality when adjusted to age, bone metastasis or heart failure. Therefore, they are essential to consider when incidental cardiac uptake is detected on bone scintigraphy. The need for pacemaker implantation was not elevated in patients with cardiac uptake.

Cardiac amyloidosis is characterized by low-molecular-weight amyloid fibrils accumulating in the myocardium (Gertz et al., 2015).The most common forms of cardiac amyloidosis are transthyretin amyloidosis (ATTR) and light-chain amyloidosis (Gertz et al., 2015).
Therefore, ATTR is an occasional incidental finding in bone scintigraphy imaged due to noncardiac indication.
In previous studies, cardiac uptake on bone scintigraphy as an incidental finding, and on patients with suspected cardiomyopathy has been associated with an elevated risk of death (Castano et al., 2016;Nitsche et al., 2022;Suomalainen et al., 2022;Uusitalo et al., 2022).However, the implications of diagnosed incidental ATTR on patient morbidity and the exact cause of death remain unclear.
Moreover, Incidental cardiac uptake on bone scintigraphy is most common amongst older men who often have multiple medical conditions that might attenuate the clinical yield of further diagnostic evaluation.Nevertheless, observed cardiac uptake might have cardiooncologic or palliative importance and, in hereditary form, might allow the early screening of the patient's family members.The detection of ATTR is often prolonged, causing a significant burden to healthcare before the correct diagnosis, and patients who receive amyloid-stabilizing medication early benefit most from the therapy (Garcia-Pavia et al., 2021;Lane et al., 2019).Therefore, additional knowledge of the natural course of incidental ATTR in oncologic patients is needed for pragmatic clinical decision-making.
In our study, we evaluate the impact of incidental cardiac uptake on bone scintigraphy on the prevalence of heart failure, arrhythmias requiring pacemaker implantation and the cause of death in elderly patients with prostate cancer.Our study provides information on the natural course of the incidental suspected ATTR on bone scintigraphy.

| Bone scintigraphy
Bone scintigrams were imaged using 99mTc-hydroxymethylenediphosphonate (99mTc-HMDP) with a standard gamma camera at 3 h postinjection.Cardiac uptake of all patients was graded using the Perugini grade, and ≥grade 2 uptake was considered positive for cardiac uptake according to the previously accepted criteria (Garcia-Pavia et al., 2021;Gillmore et al., 2016;Perugini et al., 2005).The information on bone metastases was collected using imaging reports and by reviewing available imaging data as necessary.Images were viewed and analyzed using Impax (Agfa Healthcare) and Hermes (Hermes Medical Solutions) softwares.Overall mortality due to any cause and cardiovascular mortality were used as primary outcome variables, and heart failure and pacemaker implantation as secondary clinical endpoints.The overall mortality figures also include cardiovascular deaths.The incidences of missing data were considered random and not significant for the study results.A two-tailed p-value of less than 0.05 was considered to be statistically significant.

| Patient characteristics
The study population consisted of 1426 males who underwent bone scintigraphy due to a diagnostic evaluation of prostate cancer.

| Cardiac uptake and cause of death
The causes of death in the study population are summarized in Figure 1.Of the total of 739 (52%) deaths, there were 407 (55%) deaths attributed to prostate cancer, and 149 (20%) were due to cardiovascular causes.For other reasons, non-prostate malignancies were most common (11%), followed by neurological disease (6%).
There was a nonsignificant trend towards higher overall mortality in patients with suspected ATTR (68% vs. 51%, p = 0.052).
There were 25 deaths (68%) in patients with cardiac uptake.

Cardiovascular cause of death was most common in individuals with
suspected ATTR and more common than in individuals without cardiac uptake (27% vs. 10%, p = 0.01).Most cardiac deaths (n = 10) in individuals with myocardial uptake were atherothrombotic (n = 7, 19%), of which three were caused by coronary artery disease, three by stroke and one by peripheral artery disease.The risk of atherothrombotic death was higher in patients with cardiac uptake than in other patients (19% vs. 8%, p = 0.02).Other cardiovascular reasons for death listed in individuals with suspected ATTR were dilated cardiomyopathy (n = 1, 3%), hypertensive heart and kidney disease (n = 1, 3%) and combined aortic valve disease (n = 1, 3%).
In patients with skeletal metastasis on scintigraphy (n = 467), there were 367 (79%) deaths during follow-up.The most prevalent cause of death in them was prostate cancer (n = 279, 60%), followed by cardiovascular disease (n = 42, 9%) and other malignancies (n = 25, 5%).Cardiac uptake was detected in 12 patients with metastatic bone disease but was not associated with higher overall mortality (78% vs. 83%, p = 0.69).Prostate cancer was the most common reason for mortality (n = 6, 50%) in patients with both cardiac uptake T A B L E 1 Clinical characteristics of prostate cancer patients with and without cardiac uptake on bone scintigraphy.Cardiac uptake was related to more significant overall and cardiovascular mortality, as shown in Table 2.However, as shown in Table 3, it was independently associated with cardiovascular mortality but not overall mortality.Age, bone metastasis and heart failure were independent predictors of overall and cardiovascular mortality.

| Cardiac uptake and heart failure
A diagnosis of heart failure at baseline was more common in patients with cardiac uptake than in other patients (43% vs. 14%, The spectrum of clinical events in prostate cancer patients with incidental cardiac uptake suggestive of transthyretin amyloidosis. A patient with incidental cardiac uptake on bone scintigraphy who died from prostate cancer (a).Cardiac uptake in a patient with a fatal Staphylococcus aureus sepsis resulting in infection of surgical material of previously operated spinal stenosis (b).A patient with both metastatic prostate cancer and cardiac uptake who developed atrial fibrillation and a fatal embolic stroke during follow-up (c).Myocardial uptake on bone scintigraphy in a patient who developed a new heart failure during follow-up (d).
T A B L E 2 Univariable analysis of baseline patient characteristics and cardiac uptake for overall and cardiovascular mortality prediction in prostate cancer.on the possible HF at follow-up (4 with cardiac uptake).Figure 3 shows the Kaplan−Meier analysis of the relationship between cardiac uptake and the risk of heart failure.

| Cardiac uptake and arrhythmias
As shown in Table 1, the implantation of a pacemaker for any reason was uncommon in patients with incidental cardiac uptake.
A total of 2 patients with suspected ATTR had pacemakers at baseline due to sick sinus syndrome (n = 1) and atrioventricular node ablation (n = 1).At follow-up, 1 individual with cardiac uptake received a pacemaker, but the underlying cause was not found in our patient records.No patient with cardiac uptake received ICD or CRT therapy.

| DISCUSSION
This study investigated the significance of incidental cardiac uptake suggestive of cardiac ATTR in elderly patients who underwent bone scintigraphy due to prostate cancer.Our main finding is that the risk for new heart failure was high, and cardiovascular and overall mortality were elevated in individuals with cardiac uptake.However, the overall mortality was not independently associated with cardiac uptake when adjusted to the presence of heart failure, bone metastases or age, which are essential clinical factors when evaluating patients with suspected incidental ATTR.
Indeed, patients with both heart failure or metastasis with cardiac uptake had very high mortality in our study.The risk of arrhythmia requiring pacemaker implantation was not elevated in patients with cardiac uptake.
This would suggest that unselected pacemaker population is a poor target for ATTR screening using bone scintigraphy without other clinical markers, such as left ventricular hypertrophy or heart failure with preserved ejection fraction.Our results offer additional insight into the natural history of ATTR and its cardio-oncologic consequences.

| Clinical impact of incidental ATTR
Bone scintigraphy is an accurate method for detecting ATTR, even in its subclinical phase (Glaudemans et al., 2014).Cardiac uptake has been an incidental finding in 0.4%-2% of bone scintigraphies, depending on the population studied (Al-Nahhas et al., 1995;Longhi et al., 2014;Nitsche et al., 2022;Salvalaggio et al., 2022;Suomalainen et al., 2022).When scintigraphy has been used to screen patients with aortic stenosis, heart failure or ventricular hypertrophy, the prevalence of positive studies has varied between 11% and 15% (Tini et al., 2021).At the advanced stage, ATTR causes progressive cardiomyopathy, and we did observe a higher prevalence of heart failure in patients with cardiac uptake.However, as a systemic disease, ATTR has additional cardiovascular and systemic effects beyond heart failure.First, it is associated with atrial fibrillation and other arrhythmias (Brown et al., 2022;Donnellan et al., 2020).
Amyloidosis is also a pro-thrombotic state resulting in a high risk for a subsequent (Cappelli et al., 2021;Vilches et al., 2022).Second, amyloid fibrils may infiltrate coronary vasculature and valves (Kholova, 2005).Oncologic treatment regiments might further enhance these detrimental cardiovascular effects of ATTR, while the systemic effects of amyloidosis, such as renal or hepatic failure, might impact oncologic treatments.

| Causes of death in incidental ATTR
A few previous studies have investigated the exact mode of death in patients with cardiac amyloidosis (Kharoubi et al., 2022).In a study by Kharoubi et al., there were 173 patients with a wild-type ATTR, of whom 41 died during 17 months of follow-up.Of these deaths, 59% were cardiovascular.Escher et al. included 48 wild-type ATTR patients in their analysis, of which 25% died during 5 years of followup and 80% of them due to cardiovascular reasons (Escher et al., 2020).In both studies, heart failure was the primary mode of cardiovascular death, and only 1 patient died due to cerebrovascular reasons (Escher et al., 2020;Kharoubi et al., 2022).
In agreement with previous studies, our cardiac uptake patients had a higher rate of cardiovascular death than the other patients (Escher et al., 2020;Kharoubi et al., 2022;Nitsche et al., 2022).
Atherothrombotic causes of death were the most common cause of death in our cardiac uptake patients.It is likely that ATTR, combined with cancer, confers a very high risk for an atherothrombotic event.
Cancer was the second most common reason for mortality in patients with cardiac uptake.This suggests that a pragmatic evaluation of incidental cardiac uptake is needed when planning the intensity of additional cardiac testing and possible interventions.Moreover, bone scintigraphy is positive in the early phase of cardiac ATTR, and clinically relevant ATTR cardiomyopathy might not have time to manifest during the lifetime of an elderly individual with multimorbidity.Especially in patients with both cardiac uptake and bone metastasis, cardiac mortality was rare compared to cancer deaths.

| Limitations
Our study was retrospective and based on the obtained clinical death certificate data for which an autopsy was not required.Unfortunately, the rate of autopsies in our patient population is unknown.ATTR is a rare disease in the noncardiac population, resulting in a low number of positive cases.We included only patients older than 70 years old in our study, as significant incidental cardiac uptake has been rare in the previous studies in younger patient groups and hereditary ATTR is uncommon in Finland (Longhi et al., 2014;Nitsche et al., 2022;Salvalaggio et al., 2022).
Retrospective design and noncardiac imaging indication for scintigraphy resulted in missing data on some patients.Heart failure at baseline and at the follow-up was based on clinical diagnoses on patient records and it might be possible that some cases of heart failure with preserved ejection fraction in particular might be missed in the elderly oncologic population with pragmatic end-of-the-life care.Cross-sectional single-photon emission tomography imaging was not available in our study to verify true myocardial uptake and to dismiss blood pool radioactivity as a possible cause for false positive cardiac uptake.Laboratory testing for light chain disease was not possible in our study, but we demonstrated the clinical effect of incidental cardiac uptake on prostate cancer patients.In the clinical setting, additional laboratory testing, imaging or biopsies to verify cardiac uptake as ATTR could be done by the treating physician as needed.At the time of our historical study data, the knowledge of the clinical impact of incidental cardiac uptake was poor, and treatments for cardiac ATTR were not available in Finland.Thus, imagers or referring physicians did not consider incidental cardiac uptake clinically significant, and further cardiac testing was not frequently done in this elderly population.Indeed, none of the deaths in patients with cardiac uptake were listed due to amyloidosis, suggesting a lack of autopsies in these patients.On the other hand, the risk of atherothrombotic events is elevated in both cancer and amyloidosis, and it was the leading cause of death in patients with cardiac uptake in our study.Hemodynamically significant arrhythmia events were defined as new pacemaker implantation.This might exclude some arrhythmia cases in patients who did not receive a pacemaker due to poor overall prognosis or who refused the device therapy.

| CONCLUSIONS
Incidental cardiac uptake suggestive of ATTR amyloidosis on bone scintigraphy imaged due to prostate cancer is associated with an elevated risk of heart failure and both overall and cardiovascular death.Most deaths in patients with cardiac uptake were atherothrombotic.However, cardiac uptake was not an independent predictor of overall mortality when adjusted to age, metastatic bone disease or the presence of heart failure.Therefore, they are essential to assess when managing patients with incidental cardiac uptake.The risk of pacemaker implantation was not elevated in individuals with cardiac uptake.
retrospectively screened 1426 males over 70 years old who underwent bone scintigraphy due to a previous diagnosis of prostate cancer in 2012−2018.The bone scintigraphies were imaged in three nuclear medicine departments (Jorvi Central Hospital, Kymenlaakso Central Hospital and Meilahti University Central Hospital).Exclusion criteria for the study were non-diagnostic image quality for cardiac uptake on bone scintigraphy or known diagnosis of ATTR at baseline before the imaging.Information on patient mortality was collected from obtained death certificates from the national statistical service (Statistics Finland) and was classified by the 10th International Classification of Diseases (ICD-10).Clinical patient characteristics, including diagnoses of heart failure and pacemaker implantations, were acquired from hospitals' medical records.The study was conducted in accordance with the Declaration of Helsinki, and the study protocol was approved by the local ethics committee (decision HUS/1721/2018).
Continuous variables are shown as mean ± standard deviation or median [interquartile range (IQR)] for normally distributed and skewed data.Categorical values are shown as numbers and percentages.Comparisons were done between patients with and without cardiac uptake using the χ 2 test of independence in the case of categorical variables.Normally distributed continuous variables were analyzed using the Student's t-test and variables with non-normal distribution with the Mann−Whitney U test.Cox's proportional hazards models and Kaplan−Meier curves were calculated to study the association between cardiac uptake and patient survival.The time of bone scintigraphy was the baseline of our study.

F
I G U R E 3 Kaplan−Meier analysis of overall (a) and cardiovascular mortality (b) in prostate cancer patients with and without incidental cardiac uptake.Cumulative incidence of new heart failure diagnoses during follow-up (c).p < 0.001).Heart failure at the time of positive bone scintigraphy (n = 13) resulted in a high overall and cardiovascular mortality (85% and 54%).A total of 8 patients with suspected ATTR and 138 other prostate cancer patients developed HF during the follow-up.A diagnosis of HF at follow-up was more common in patients with suspected ATTR (47% vs. 15%, p < 0.001).The median time to HF diagnosis in individuals with cardiac uptake was 5 years (IQR: 1-6 years).A total of 13 individuals with a diagnosis of HF in patient records were excluded from the analyses since the date of the HF diagnosis was unknown (3 patients with cardiac uptake).In addition, 281 individuals were excluded due to no information