Thromboembolic events after high‐intensity training during cisplatin‐based chemotherapy for testicular cancer: Case reports and review of the literature

The randomized “Testicular cancer and Aerobic and Strength Training trial” (TAST‐trial) aimed to evaluate the effect of high‐intensity interval training (HIIT) on cardiorespiratory fitness during cisplatin‐based chemotherapy (CBCT) for testicular cancer (TC). Here, we report on an unexpected high number of thromboembolic (TE) events among patients randomized to the intervention arm, and on a review of the literature on TE events in TC patients undergoing CBCT. Patients aged 18 to 60 years with a diagnosis of metastatic germ cell TC, planned for 3 to 4 CBCT cycles, were randomized to a 9 to 12 weeks exercise intervention, or to a single lifestyle counseling session. The exercise intervention included two weekly HIIT sessions, each with 2 to 4 intervals of 2 to 4 minutes at 85% to 95% of peak heart rate. The study was prematurely discontinued after inclusion of 19 of the planned 94 patients, with nine patients randomized to the intervention arm and 10 to the control arm. Three patients in the intervention arm developed TE complications; two with pulmonary embolism and one with myocardial infarction. All three patients had clinical stage IIA TC. No TE complications were observed among patients in the control arm. Our observations indicate that high‐intensity aerobic training during CBCT might increase the risk of TE events in TC patients, leading to premature closure of the TAST‐trial.

prematurely discontinued after inclusion of 19 of the planned 94 patients, with nine patients randomized to the intervention arm and 10 to the control arm. Three patients in the intervention arm developed TE complications; two with pulmonary embolism and one with myocardial infarction. All three patients had clinical stage IIA TC. No TE complications were observed among patients in the control arm. Our observations indicate that high-intensity aerobic training during CBCT might increase the risk of TE events in TC patients, leading to premature closure of the TAST-trial.

| INTRODUCTION
The potential benefits of exercise training during and after cancer treatment have increasingly gained interest. Current evidence suggests that exercise is safe and effective to maintain or improve physical fitness and patient-reported outcomes both during and after treatment. [1][2][3] Cancer patients and survivors are therefore generally recommended to avoid inactivity and follow the public guidelines for physical activity if feasible. [4][5][6] However, the optimal intensity of exercise during cancer treatment remains unclear; particularly, the efficacy, feasibility and safety of high-intensity training (HIT) across subgroups of cancer patients. 7 Previous randomized controlled trials (RCTs) examining effects and safety of HIT during chemotherapy have demonstrated beneficial effects and few adverse events (AEs). [8][9][10][11][12] Notably, only a small minority of patients in these studies were treated with cisplatin.
Cisplatin-based chemotherapy (CBCT) is standard treatment for metastatic germ cell testicular cancer (TC). 13 During CBCT, TC patients frequently experience reduced cardiorespiratory fitness (CRF) and muscle strength. Furthermore, TC survivors who have received CBCT are at risk of chronic fatigue and development of metabolic syndrome. 14,15 Given the risk of acute and long-term AEs after treatment of metastatic TC, identification of risk-reducing interventions is of high relevance. To the best of our knowledge, only one study has examined the effects of exercise during chemotherapy for TC, suggesting that high-intensity strength training was safe. 16 In the "Testicular cancer and Aerobic and Strength Training trial" (TAST-trial), we aimed to evaluate the effects of high-intensity interval training (HIIT) on CRF in TC patients during CBCT. Here, we report on the unexpected high number of thromboembolic (TE) events among the patients randomized to the intervention.

| Study design and patients
The TAST-trial was a two-arm (1:1 ratio) national multicenter RCT, comparing change in CRF measured by peak oxygen uptake (VO 2peak ) in TC patients who during CBCT underwent a training program including HIIT, to controls who received a single lifestyle counseling session. The randomization was computerized in an equal allocation and the patients were stratified by study center.
Patients were recruited at four university hospitals in Norway:

| Study assessments
All participants underwent the same assessments before and after the intervention period. The primary outcome was VO 2peak measured

What's new?
Exercise is widely acknowledged to be beneficial for cancer patients, but certain subgroups may experience adverse effects. Patients with testicular cancer who take cisplatinbased chemotherapy often experience reduced cardiovascular function. In this study, the authors observed an increase in thromboembolic events among testicular cancer patients taking cisplatin who engaged in HIIT twice a week. In a randomized trial, 3 out of 9 patients in the exercise arm experienced TE events, including myocardial infarction and pulmonary embolism, compared with 0 of 10 patients in the control arm.
The study was cut short due to the increased risk.
during a CPET using a continuous graded exercise protocol on a treadmill until exhaustion. Peak heart rate (HR peak ) assessed by 12-leads electrocardiography (ECG) was also measured during the CPET. If the rest and exercise-ECGs were normal and no cardiac symptoms occurred, no further cardiac examinations were performed. For safety reasons, pulmonary function, blood pressure and saturation were also measured before, during and after CPET, all under supervision of an exercise physiologist and a physician. Other assessments included, muscle strength tests, dual-energy X-ray absorptiometry (DXA) scan, routine blood tests and questionnaires.
After the discontinuation of the TAST-trial, the cases were assessed with regard to individual susceptibility for TE complications.

| Intervention arm
Because CRF was the primary outcome of the trial, we composed an intervention that emphasized high-intensity aerobic exercise. The intervention included two one-to-one supervised sessions per week for 9 or 12 weeks depending on the number of chemotherapy cycles.
Walking uphill on a treadmill was the primary choice of exercise, but ergometer-cycling, rowing and out-door walking were possible alternatives. Each session consisted of 10 minutes warm-up at 60% to 70% of HR peak , followed by HIIT; that is, 2 to 4 intervals of 2 to 4 minutes at 85% to 95% of HR peak, (high-intensity zone); separated by 2 minutes' active recovery, and followed by 10 minutes' cool-down ( Figure 1). Thereafter, an optional 15 minutes strength training was performed, depending on the patient's energy level. HR peak obtained during the pre-intervention CPET was used to calculate the training zones. The HR and the Borg Rating of Perceived Exertion Scale 17 were registered each minute during the HIIT. If the patients felt unwell before or during a session, the physiotherapists/personal trainers were instructed to make individual adaptations regarding the intensity, duration and/or number of intervals. If the patient's condition was not compatible with HIIT, the planned session was postponed or canceled.

| Control arm
During the first chemotherapy cycle, patients in the control arm received a 30-minute counseling session on general lifestyle recommendations.

| Sample size calculation
Based on our experience from a pilot study, 18 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29  Minutes F I G U R E 1 Typical session including 10 minutes warm-up, four intervals of high-intensity (85%-95% of HR peak ) physical exercise separated by active recovery, and 10 minutes cool-down. HR, heart rate; HR peak , peak heart rate (beatÁmin −1 ) [Color figure can be viewed at wileyonlinelibrary.com] group and 9 mL/kg/min in the exercise group, with a SD of 5 mL/kg/ min for both groups. Upfront sample size calculations showed that we needed 47 patients in each group to detect a mean group difference in change of VO 2 peak during chemotherapy of 5 mL/kg/min with a SD of 7 mL/kg/min (two-sided significance level of 5%, power of 90% and 10% dropout).

T A B L E 1
Baseline characteristics of the patients in the intervention-and control arm, and of each case who developed a thromboembolic event   i Psychological distress (n = 2), muscle and skeletal disease (n = 2) and chronic lung disease (n = 1). j Hypertension (n = 1), muscle and skeletal pain/disorder (n = 2), mild/moderate asthma (n = 1) and epilepsy (n = 1). k Psychological distress. l Muscle and skeletal pain/disorder. pain and dyspnea from day seven of the second BEP cycle, which he had not reported to health professionals. He received dalteparin subcutaneously for 10 months, thereafter warfarin for 4 months.

Potential risk factors for venous TE
Re-evaluation of the prechemotherapy CT scan did not reveal venous thrombosis or pulmonary embolism. This patient had no hereditary factors for venous TE events, and all coagulation analyses were within reference ranges. of thromboembolic rather than atherosclerotic origin in TC patients during CBCT has been described previously. 20 He was given ticagrelor for 3 months and long-term acetylsalicylic acid and atorvastatin.

Potential risk factors for arterial TE
He had hereditary risk factors for cardiovascular disease, as one parent had angina pectoris, and several family members had hypercholesterolemia. All coagulation analyses were within reference ranges. His body mass index was 32 kg/m 2 . He had no signs of hypertension (blood pressure 110/75 mmHg) or hyperglycemia. He was a never-smoker. Although within reference ranges, the lipid profile at baseline was unfavorable (Table 1). Taken together, this patient had a modest increased risk of arterial TE.
All three cases completed three cycles of BEP as planned, achieving durable complete remissions.

| DISCUSSION
Three of nine patients in the intervention arm experienced a TE event, that is, pulmonary embolism and myocardial infarction. Since the risk of pulmonary embolism or myocardial infarction are expected to be low in patients with TC during or shortly after CBCT, our observations indicated a substantially higher rate than reported in the literature (33% vs 0-15%). Pretreatment, none of the cases had any TC-specific risk factors for TE events, such as large abdominal lymph nodes, elevated LDH or central venous access. In accordance with national guidelines for treatment of TC in Norway, they did not receive primary thromboprophylaxis. Case 1 and 3 had predisposing factors for venous and arterial TE events, respectively. 20 Although we are fully aware that the TE events in the intervention arm might have been a play of chance, we find it hard to ignore that the HIIT might have contributed to the unexpected high number of TE events. Proposed mechanisms for possible interactions between CBCT and high-intensity aerobic exercise that can potentiate the risk of TE events are described after a review of the literature.

| TE events during CBCT-review of the literature
Within the three main concepts; testicular neoplasms, cisplatin and thromboembolism; MeSH terms with variations were searched in titles, abstracts and author keywords in MEDLINE (Ovid) and Embase

T A B L E 2
Number of intervals and minutes when the cases reached high-intensity training zone (85%-95% of peak heart rate) for each session T   (Table 3). Studies limited to venous access-associated thrombosis were excluded. The selection process is further detailed in Figure S2.

| The association of HIT and TE risk
Blood is known to be hypercoagulable immediately after strenuous exercise, mainly due to an increased level of coagulation factor VIII. 43 Studies have shown that HIT sessions are followed by a transient increase in platelet activation and aggregation. [44][45][46] The increase in factor VIII and the degree of platelet activation and aggregation after exercise are associated with the intensity of the exercise. 47 them more prone to TE events. It is possible that exercise programs with lower intensity and more gradual increase in intensity might be more favorable than the HIIT program in the TAST-trial.

| Limitations
The small number of included patients in the TAST-trial is an obvious limitation. Possibly, this is an accidental observation unrelated to the HIIT, reflected in the wide CI. Furthermore, our observations are after high-intensity aerobic exercise, thus not representative for physical activity with low-and moderate intensity or strength training. We are unable to estimate an exact cut-off for the training intensity regarding the risk of TE events. Our observation is also limited to HIT during CBCT, not to training after completing CBCT. Future research on TC and exercise training may consider exercise protocols with lower intensity during CBCT or HIT interventions after completion of CBCT.

| CONCLUSION
It is well known that TC patients are at risk of TE events during CBCT.
Two of the three cases with TE events had risk factors for such events. Our study raises the possibility that HIIT during CBCT adds to CBCT-induced hypercoagulability.

CONFLICT OF INTEREST
Prof Wisloff has received funding for work with Varicella and Herpes Zoster vaccine from MSD, not relevant for this article. The other authors declare no potential conflicts of interest.

DATA AVAILABILITY STATEMENT
The data that support the findings of our study are available on request from the corresponding author, and with permission from Regional Com-

ETHICS STATEMENT
The TAST-trial was approved by the Regional Committee for Medical and Health Research Ethics (2014/1169/REC South-East) and registered in ClinicalTrial.gov (NCT02577172). All participants signed an informed consent before inclusion, and the three cases have read this report and provided a written consent for publication.