Impact of maintenance treatment on male gonadal function in patients treated for localized rhabdomyosarcoma in RMS2005 trial in France

Abstract

Background

Maintenance treatment with vinorelbine and oral cyclophosphamide (oral‐CPM) improves outcome of nonmetastatic high‐risk (HR) and very‐high risk (VHR) rhabdomyosarcoma (RMS) patients. However, gonadal toxicity of maintenance was not yet investigated.

Methods

The authors focused their analysis on male gonadal toxicity in HR/VHR groups of RMS2005 trial, in France. In the HR group, patients were randomized to receive or not receive 6 months of maintenance (after nine or four IVADo] + five IVA). In the VHR group, patients received 6 months of maintenance (after four IVADo + five IVA). Exocrine gonadal dysfunction (EGD) was defined as followed: follicle‐stimulating hormone level >10 IU/L and/or inhibin‐B <80 pg/mL and/or oligoasthenozoospermia/azoospermia.

Results

Among 86 eligible 5‐year RMS survivors ≥12 years old, 49 had available gonadal evaluation (median age at diagnosis = 6.4 years, median age at evaluation =  18.7 years, 41 HR/8 VHR). Twenty‐six (53%) received oral‐CPM (median cumulative dose = 4.2 g/m², range = 0.7–9.0). EGD was reported in 18 of 49 (37%). Exposure to oral‐CPM (odds ratio [OR], 5.45; 95% confidence interval [CI], 1.30–22.92, p = .021) and older age at diagnosis (compared to 0–5 years, OR_{5–10 years}, 9.61; 95% CI, 1.49–62.15 and OR_{>10 years}, 14.10; 95% CI, 1.89–105.33, p = .025) were significantly associated with EGD. Higher cumulative dose of oral‐CPM (>4.5 g/m²) tended to be more toxic (compared to nonexposure, OR_{≤4.5 g/m2}, 2.95; 95% CI, 0.78–11.09 and OR_{>4.5 g/m2}, 7.20; 95% CI, 1.01–51.39, p = .094).

Conclusions

Oral‐CPM induces additional gonadal damage to the ifosfamide‐based induction regimen. Fertility preservation could be considered in patients exposed to maintenance, especially those >5 years old and exposed to ≥12 months of oral‐CPM.

Short abstract

In this analysis on male gonadal toxicity in high‐risk and very‐high risk groups of patients with rhabdomyosarcoma treated in RMS2005 trial, in France, exposure to maintenance treatment with oral cyclophosphamide and older age at diagnosis were significantly associated with exocrine gonadal dysfunction. Fertility preservation could be considered in patients exposed to maintenance, especially those >5 years old and exposed to ≥12 months of oral cyclophosphamide.

INTRODUCTION

Rhabdomyosarcoma (RMS) is the most common soft tissue sarcoma in childhood. ¹ It represents 6%‐7% of cancers before the age of 20, with approximately 120 cases in France each year. Five‐year event‐free (EFS) and overall (OS) survivals are approximately 70% and 80%, respectively, for patients with nonmetastatic RMS. ² In the international prospective randomized phase 3 trial RMS2005, adding six cycles of maintenance chemotherapy with intravenous vinorelbine and low‐dose oral cyclophosphamide to the standard therapy based on nine cycles of ifosfamide, vincristine, dactinomycin with or without doxorubicin (IVA ± Do) and local treatment (surgery, radiotherapy, or both) in patients with nonmetastatic high‐risk (HR) RMS, led to a significant improvement of 5‐year OS, from 73.7% to 86.5% (HR, 0.52; 95% confidence interval [CI], 0.32–0.86; p = .0097). ³ In patients with very high‐risk (VHR) localized RMS (alveolar histotype with positive regional RMS and nodes), six cycles of vinorelbine‐cyclophosphamide was added to historical backbone chemotherapy regimen (four IVADo + five IVA) resulting in 5‐year EFS of 50.1% (vs. 29%–57% in historical controls). ³ , ⁴ To increase survival, the ongoing international randomized phase 3 trial EpSSG FAR‐RMS (ClinicalTrials.gov identifier: NCT04625907) compares 6 months versus 12 months of vinorelbine‐cyclophosphamide maintenance treatment in patients with HR RMS and 12 months versus 24 months in VHR disease. ⁵

If the acute toxicity of such regimen is well described, no data concerning late effects were available at time of trial results publication, especially concerning gonadal function. However, the gonadal toxicity of the cumulative dose of alkylating agents is well known, especially with conventional discontinuous dose of intravenous cyclophosphamide >6 g/m² for males. ⁶ , ⁷ , ⁸ , ⁹ , ¹⁰ Ifosfamide gonadal toxicity seems to be inferior to that of intravenous cyclophosphamide, which questions the difference between hematological and gonadal toxicities equivalent doses. ⁷ , ¹⁰ At that time, the cyclophosphamide equivalent dose (CED) considers only hematological toxicity for equivalence estimation, leading to a ratio of 4 between ifosfamide and cyclophosphamide. ¹¹ However, this ratio is questionable in the context of gonadal toxicity and could be investigated in the same manner as a doxorubicin‐equivalent dose for cardiac toxicity. ¹² , ¹³ Concerning oral cyclophosphamide, there is no knowledge of any difference of gonadal toxicity between continuous low‐dose oral and push high‐dose intravenous administration. However, gonadal toxicity of low‐dose continuous cyclophosphamide has not yet been investigated, and no studies are available concerning the potential additive toxicity of this oral alkylating agent after ifosfamide based‐regimen.

In this analysis, we aimed to study the impact of low‐dose oral cyclophosphamide in maintenance treatment on male gonadal function in patients with HR and VHR nonmetastatic RMS treated in the RMS2005 trial in France.

MATERIALS AND METHODS

Study population and data collection

The European Pediatric Soft Tissue Sarcoma Study Group (EpSSG) RMS2005 trial addressed the treatment of children, adolescents, and young adults (<21 years old) presenting with nonmetastatic RMS. The protocol contained a randomized trial for patients with HR disease investigating the role of doxorubicin dose intensity and maintenance chemotherapy and observational studies for patients categorized in other risk groups. ³ The HR group included patients in incomplete resection with microscopic or macroscopic residual tumor (IRS groups II or III, respectively) with nonalveolar histology but unfavorable sites (other than orbit, genito‐urinary nonbladder prostate, and nonparameningeal head and neck), size (>5 cm), or age (≥10 years); patients with nonalveolar histology and nodal involvement; and patients with localized alveolar histology. The VHR group included patients with alveolar histology and node involvement, regardless of the other risk factors. Patients in HR group were randomized to 1) receive nine IVA (ifosfamide, vincristine and dactinomycin) versus four IVADo + five IVA in induction, and 2) to stop treatment or to continue with low‐dose maintenance therapy with a combination of vinorelbine (25 mg/m² on day 1, day 8, and day 15) and continuous low‐dose cyclophosphamide (25 mg/m²/day on days 1–28) for six cycles. Patients in the VHR group should have received four IVADo cycles followed by five IVA blocks and low‐dose cyclophosphamide and vinorelbine maintenance treatment for six cycles, however, protocol deviations were observed resulting in administration of 12 cycles. The total planned cumulative doses of chemotherapy were 54 g/m² of ifosfamide, 240 mg/m² of doxorubicin, and 4.2 g/m² (six cycles) of oral cyclophosphamide. ³ In the analysis of patients treated within RMS 2005 study, the 5‐year EFS in fusion‐positive, node‐positive patients (VHR patients in this study) was 43% (95% CI, 30–56), showing the need for improved treatments for patients with fusion‐positive disease with lymph node involvement. ⁴ This may explain why some investigators chose to give a priori 12 months of maintenance treatment. Notably, in view of the similarly poor outcomes for fusion‐positive/node‐positive RMS and metastatic RMS, these groups are combined in current FAR‐RMS to give a newly defined VHR group.

Male patients treated for nonmetastatic RMS in HR and VHR groups in the RMS2005 trial in France between 2005 and 2016, alive in first complete remission (CR) without second malignant neoplasms (SMN) in the 5 years post‐diagnosis with at least 5 years follow‐up since diagnosis, and an attained age at last follow‐up of at least 12 years were included in this analysis.

Demographic information, tumor characteristics, cancer treatments (including surgical methods, chemotherapy cumulative doses, and radiation therapy fields and doses) were extracted from RMS2005 database. Missing data were completed thanks to each oncological center. Mortality and late morbidity data were updated in 2023. Consent for data collection was obtained from patients and for minors from parents or guardians according to national research ethics requirements at time of trial enrollment.

Gonadal function was assessed by follicle‐stimulating hormone (FSH), luteinizing hormone (LH), testosterone, inhibin B measurements, and by sperm count analysis according to feasibility. Exocrine gonadal dysfunction was defined as FSH >10 IU/L and/or inhibin B <80 pg/mL and/or oligoasthenozoospermia and/or azoospermia. ¹⁴ Endocrine gonadal dysfunction was defined as LH >8.6 IU/L and testosterone <1.4 ng/mL and/or hormonal replacement therapy with testosterone. For patients with iterative gonadal function assessments, the last available evaluation was used. No data on gonadal function after induction were available. All measurements were performed after the end of maintenance treatment.

Statistical analysis

OS was defined as time between diagnosis and death from any cause. EFS was defined as time between diagnosis and progression/relapse, SMN, or death from any cause, whichever occurred first. Survival curves were estimated using the Kaplan–Meier method and 95% CI computed with the Rothman method. Student t‐tests, χ² tests, and log‐rank tests were used to compare means, percentages, and survival curves, respectively, in patients with or without gonadic function assessment.

We performed risk factor analysis for exocrine gonadal dysfunction. Odds ratios (ORs) with 95% Wald CIs were estimated by univariate logistic regression models for the following clinical characteristics and cancer treatment variables: age at diagnosis (0–5 years, 5–10 years, and >10 years), genito‐urinary primary tumors (yes/no), pubertal status at diagnosis (pre‐pubertal vs. intra‐ or post‐pubertal), oral cyclophosphamide cumulative dose (yes/no or by categories 0/≤4.5 g/m²/>4.5 g/m², with 4.5 g/m² as a cutoff to separate patients who received 6 months of oral maintenance from those exposed to longer duration), orchiectomy (yes/no), pituitary gland radiation dose (≤24 Gy vs. >24 Gy), and testicular radiation dose (≤1 Gy vs. >1 Gy). Variables significantly associated with exocrine gonadal dysfunction at threshold 0.05 were included in multivariable regression models. All tests were two‐sided with α set to 0.05. Statistical analyses were performed with SAS 9.4 (SAS Institute Inc, Cary, North Carolina) and R (http://www.r‐project.org/) software.

RESULTS

Among 497 patients with nonmetastatic RMS treated in France in the RMS2005 trial, 174 males were included in HR and VHR groups. Fifty‐seven patients were excluded, due to events (progression/relapse, n = 50; SMN, n = 1; and death as first event, n = 2) or loss to follow‐up (n = 4) in the 5 years post‐diagnosis. A total of 117 male patients were alive in first complete response, with at least 5 years of follow‐up since diagnosis. We excluded 31 patients less than 12 years old at last follow‐up. Overall, 86 5‐year RMS survivors with an attained age of ≥12 years were eligible, 75 in HR group and 11 in VHR (Figure 1). Patients’ characteristics are presented in Table 1. Median age at diagnosis was 6.3 years (range, 0.1–21.8), and 41 patients had embryonal RMS (47.7%). Forty patients received maintenance treatment with oral cyclophosphamide at a median cumulative dose of 4.2 g/m² (0.7–9). A local treatment was performed in 83 patients (surgery only, n = 8; radiation therapy only, n = 20; and both, n = 55). With a median follow‐up of 10.6 years (range, 5.8–16.4), we did not observe any event or death in these 86 patients.

FIGURE 1.

Flow chart. CR indicates complete response; HR, high‐risk; LR, low‐risk; RMS, rhabdomyosarcoma; S, standard‐risk; VHR, very‐high‐risk; y, years.

TABLE 1.

Clinical, therapeutic, and follow‐up characteristics of the 5‐year RMS survivors with an attained age ≥12 years treated in RMS‐2005 trial in France and eligible for the male gonadal toxicity analysis (n = 86).

Patient and tumor characteristics	All eligible patients (N = 86)	Patients with gonadal function assessment (n = 49)	Patients without gonadal function assessment (n = 37)	p
Age at diagnosis (years)				.81
Mean [SD]	8.3 [5.4]	8.0 [5.5]	8.3 [5.3]
Median [range]	6.3 [0.1–21.8]	6.4 [0.1–20.7]	6.3 [1.1–21.8]
Primary tumor location, No. (%)				.09 a
Head/neck	41 (47.7)	24 (49.0)	17 (45.9)
Bladder/prostate	19 (22.1)	14 (28.6)	5 (13.5)
Paratesticular	7 (8.1)	4 (8.2)	3 (8.1)
Abdomen/pelvis	9 (10.5)	3 (6.1)	6 (16.2)
Limb	6 (7.0)	1 (2.0)	5 (13.5)
Thorax	4 (4.7)	3 (6.1)	1 (2.7)
RMS2005 risk group, No. (%)				.26
High risk	75 (87.2)	41 (83.7)	34 (91.9)
Very high risk	11 (12.8)	8 (16.3)	3 (8.1)
Histology, No. (%)				.05
Embryonal	57 (66.3)	37 (75.5)	20 (54.1)
Alveolar	28 (32.6)	12 (24.5)	16 (43.2)
Unknown	1 (1.1)	0 (0.0)	1 (2.7)
Local treatment, No. (%)
Surgery only	83 (96.5)	48 (98.0)	35 (94.6)	.40
Radiotherapy only	8 (9.3)	3 (6.1)	5 (13.5)	.21
Surgery + radiotherapy	20 (23.3)	12 (24.5)	8 (21.6)	.74
None	55 (64.0)	33 (67.3)	22 (59.5)	.67
Unknown	1 (1.2)	1 (2.0)	0 (0.0)
	2 (2.3)	0 (0.0)	2 (5.4)
Chemotherapy
Ifosfamide
No. of patients exposed	85 (98.8%)	48 (98.0%)	37 (100%)	.38
Median cumulative dose (g/m2) [range]	54 [24–59]	54 [24–59]	54 [36–56]
Dose missing	20	0	20
Intravenous CPM
No. of patients exposed	10	6 (12.2%)	4 (10.8%)	.38
Median cumulative dose (g/m2) [range]	3 [1–12]	3 [1–12]	1.3 [1–3.5]
Dose missing	0	0	0
Doxorubicin
No. of patients exposed	42 (48.8%)	25 (51.0%)	17 (45.9%)	.51
Median cumulative dose (mg/m2) [range]	240 [75–240]	240 [75–240]	240 [180–240]
Dose missing	9	0	9
Oral CPM
No. of patients exposed	40 (46.5%)	26 (53.1%)	14 (37.8%)	.16
Median cumulative dose (g/m2) [range]	4.2 [0.7–9]	4.2 [0.7–9]	4.2 [2.3–7.5]
Dose missing	5	0	5
Follow‐up after diagnosis
Mean [SD]	10.7 [3.1]	11.7 [2.7]	9.5 [2.6]	<.01
Median [range]	10.6 [5.8–16.4]	11.7 [6.4–16.4]	9.3 [5.8–15.8]
Age at last follow‐up (years)
Mean [SD]	18.9 [5.4]	19.7 [5.1]	17.8 [4.8]	.09
Median [range]	18.1 [12.0–33.9]	18.7 [12.0–33.9]	16.4 [12.0–30.4]
Second event after 5 years post‐diagnosis (relapse, second malignancy, or death), No. (%)	0 (0.0)	0 (0.0)	0 (0.0)
10‐year EFS	100%	100%	100%
10‐year OS	100%	100%	100%

Note: Characteristics of patients with or without gonadal function assessments were compared.

Abbreviations: CPM, cyclophosphamide; EFS, event‐free survival; HR, high‐risk; LTFU, long‐term follow‐up; OS, overall survival; RMS, rhabdomyosarcoma; VHR, very‐high‐risk.

^a For statistical analysis, patients were divided into three groups: head/neck, bladder/prostate/paratesticular, and other.

Gonadal function evaluations were available for 49 patients who consequently could be included in this male gonadal function analysis (Figure 1). Forty‐one patients were treated in HR group and eight patients were treated in VHR. A detailed flow‐chart is represented in Figure S1. The clinical and therapeutic characteristics of these patients were similar to those without gonadal function evaluation but included patients that had longer follow‐up and were older at last follow‐up (Table 1). Four patients had paratesticular primary treated with orchidectomy and external beam radiation therapy (RT). Fourteen patients had bladder prostate RMS treated with surgery in three cases, RT in three cases, and a combination of surgery and RT in eight cases. Among them, 14 patients had external beam RT and eight patients had brachytherapy resulting in testicular dose >1 Gy in 11 patients. Twenty‐four patients had head and neck RMS treated with RT resulting in pituitary gland radiation >24 Gy in 23 cases. A total of 48 of 49 patients were exposed to ifosfamide (median cumulative dose = 54 g/m²; range, 24–59), six patients (12%) were exposed to intravenous cyclophosphamide (median cumulative dose = 3 g/m²; range, 1–12, instead of ifosfamide due to disease‐related acute renal injury at diagnosis or secondary ifosfamide‐related nephrotoxicity in induction), and 26 patients (53%) had a maintenance treatment with oral cyclophosphamide (median cumulative dose = 4.2 g/m²; range, 0.7–9). The clinical and therapeutic characteristics of these 49 patients according to exposure to maintenance therapy and to the cumulative dose of oral cyclophosphamide are reported in Table 2.

TABLE 2.

Clinical and therapeutic characteristics of patients included in the male gonadal toxicity analysis according to maintenance therapy and to the cumulative dose of oral CPM.

Patients’ characteristics	No maintenance treatment (N = 23)	Maintenance treatment (N = 26)
		All exposed patients (N = 26)	Oral‐CPM ≤4.5 g/m2 (N = 20)	Oral‐CPM >4.5 g/m2 (N = 6)
Age at diagnosis, median (years)	7.3 [1 months–20.7 years]	6.3 [10.8 months–20.7 years]	5.8 [11.1 months–19.5 years]	9.6 [4.3 years–20.7 years]
Follow‐up duration, median (years)	12.0 [6.4–16.1]	11.5 [6.7–16.4]	11.5 [6.7–16.4]	11.7 [7.0–14.4]
Age at last follow‐up, median (years)	17.9 [11.5–29.6]	18.7 [12.8–33.9]	18.7 [12.8–33.9]	20.9 [18.3–27.8]
Primary tumor location, No. (%)
Head/neck	13 (56.5)	11 (42.3)	9 (45)	2 (33.3)
Bladder/prostate	7 (30.4)	7 (26.9)	4 (20.0)	3 (50)
Paratesticular	2 (8.7)	2 (7.7)	2 (10.0)	0 (0)
Abdomen/pelvis	0	3 (11.5)	2 (10.0)	1 (16.7)
Thorax	1 (4.3)	2 (7.7)	2 (10.0)	0 (0)
Limb	0	1 (3.8)	1 (5.0)	0 (0)
Risk group, No. (%)
High risk		18 (69.2)	14 (70.0)	4 (66.7)
Very high risk	23 (100)	8 (30.8)	6 (30.0)	2 (33.3)
Primary tumor surgery, No. (%)	15/23 (65.2)	21/26 (80.8)	18/20 (90.0)	3/6 (50.0)
With orchiectomy	2/23 (8.7)	2/26 (7.7)	2/20 (10.0)	0/6 (0)
Primary tumor radiotherapy
Type, No. (%)	21/23 (91.3)	24/26 (92.3)	18/20 (90.0)	6/6(100)
External beam RT	16/21 (76.2)	22/24 (91.7)	16/18 (88.9)	5/6 (83.3)
Brachytherapy	6/21 (28.6)	2/24 (8.3)	1/18 (5.6)	1/6 (16.7)
Pituitary gland dose >24 Gy	12/21 (57.1%	11/24 (45.8)	9/18 (50.0)	2/6 (33.3)
Testicular dose >1 Gy	6/21 (28.6)	5/24 (20.8)	4/28 (22.2)	1/6 (16.7)
Chemotherapy
Ifosfamide
No. of patients exposed	22/23 (95.7%)	26/26 (100%)	20/20 (100%)	6/6 (100%)
Median cumulative dose (g/m2) [range]	54 [24–55.4]	54 [24–59]	54 [24–55.6]	54 [41.6–59]
Intravenous CPM
No. of patients exposed	2/23 (8.7%)	4/26 (15.4%)	2/20 (10.0%)	2/6 (33.3%)
Median cumulative dose (g/m2) [range]	4.5 [1.4–7.5]	3 [1–12]	6.5 [1–12]	3 [3–3]
Oral CPM
No. of patients exposed	0/23 (0%)	26/26 (100%)	20/20 (100%)	6/6 (100%)
Median cumulative dose (g/m2) [range]		4.2 [0.7–9]	4.1 [0.7–4.5]	8.7 [6.8–9]
Fertility preservation
No. of patients with fertility preservation	3/23 (13.0%)	3/26 (11.5%)	2/20 (10.0%)	1/6 a (16.7%)
Type
Unilateral testicular transposition	0	1	1	0
Bilateral testicular transposition	0	1	0	1
Testicular tissue banking	0	0	0	0
Sperm sample cryopreservation	3	2	1	1

Abbreviations: CPM, cyclophosphamide; RT, radiotherapy.

^a One patient had fertility preservation with both sperm sample cryopreservation and unilateral testicular transposition.

Concerning pubertal development, 36 patients were pre‐pubertal at diagnosis; among them, 35 developed a spontaneous puberty (one was still pre‐pubertal at the age of 12.6 years at last follow‐up). At last follow‐up, FSH/LH dosages were available for 48 patients, testosterone for 44 patients, inhibin B for 31 patients, and sperm count analysis for three patients. Results of gonadal function assessment according to exposure to maintenance therapy and to the cumulative dose of oral cyclophosphamide are reported in Table 3.

TABLE 3.

Results of gonadal toxicity function assessment according to exposure to maintenance therapy and to the cumulative dose of oral CPM.

Patients’ characteristics	No maintenance treatment (N = 23)	Maintenance treatment (N = 26)
		All exposed patients (N = 26)	Oral‐CPM ≤4.5 g/m2 (N = 20)	Oral‐CPM >4.5 g/m2 (N = 6)
Pubertal status
Before treatment, No. (%)
Pre‐pubertal	17 (73.9)	19 (73.1)	16 (80.0)	3 (50)
Intra‐pubertal	1 (4.3)	2 (7.7)	1 (5.0)	1 (16.7)
Post‐pubertal	5 (21.7)	5 (19.2)	3 (15.0)	2 (33.3)
At last clinical evaluation, No. (%)
Pre‐pubertal	1 (4.3)	0	0 (0)	0 (0)
Intra‐pubertal	3 (13.0)	5 (19.2)	5 (25.0)	0 (0)
Post‐pubertal	19 (82.6)	21 (80.8)	15 (75.0)	6 (100)
Hormonal dosages
Median age at time of hormonal dosage (years)	17.2 [11.5–29.6]	18.3 [11.5–27.7]	17.1 [11.5–27.0]	20.0 [18.3–27.7]
Median interval between diagnosis and hormonal dosage (years)	11.4 [2.8–15.9]	11.1 [3.1–15.9]	10.6 [3.1–15.9]	11.6 [6.4–14.4]
FSH
Median [range] (IU/L)	4.9 [0.8–20.2]	6.5 [1.4–54.2]	5.2 [1.4–54.2]	9.5 [1.4–15.1]
No. of patients with abnormal dosage a	4/23 (17.4%)	7/25 (28.0%)	5/19 (26.3%)	2/6 (33.3%)
LH
Median [range] (IU/L)	3.9 [0.4–9.9]	4.0 [1.0–34.1]	4.4 [1.0–34.1]	3.6 [1.6–5.9]
No. of patients with abnormal dosage a	1/23 (4.3%)	3/25 (12.0%)	3/19 (15.8%)	0/6 (0.0%)
Testosterone
Median [range] (ng/mL)	5.5 [0.3–18]	5.2 [0.2–10.9] (23)	5.2 [0.2–8.3]	5.2 [1.4–10.9]
No. of patients with abnormal dosage a	2/23 (8.6%)	1/23 (4.3%)	1/17 (5.8%)	0/5 (0.0%)
Inhibin B
Median [range] (mg/dL)	156 [17–355]	105.5 [8–652]	133 [8–652]	26.5 [15.0–174] (4)
No. of patients with abnormal dosage a	4/13 (30.7%)	9/18 (50%)	6/14 (42.8%)	3/4 (75.0%)
Sperm count analysis
Median age at time of semen analysis (years)	1/23 (4.3%)	2/26 (7.7%)	1/20 (5.0%)	1/6 (16.7%)
Time between diagnosis and semen analysis (years)	16.5	20.2 [18.9–21.5]	18.9	21.5
Results	10.3	4.7 [4.4–4.9]	4.4	4.9
Normal	1 (100%)	0 (0%)	0 (0%)	0 (0%)
Azoospermia	0 (0%)	2 (100%)	1 (100%)	1 (100%)
Exocrine gonadal dysfunction b	5/23 (21.7%)	13/26 (50%)	9/20 (45%)	4/6 (66.7%)
Endocrine gonadal dysfunction c	1/23 (4.3%)	1/26 (3.8%)	1/20 (5.0%)	0/6 (0%)

Abbreviations: CPM, cyclophosphamide; FSH, follicle‐stimulating hormone; LH, luteinizing hormone.

^a Reported on number of available dosages.

^b Exocrine gonadal dysfunction was defined as FSH >10 IU/L and/or inhibin B <80 mg/dL and/or oligoasthenozoospermia and/or azoospermia.

^c Endocrine gonadal dysfunction was defined as LH >8.6 IU/L and testosterone <1.4 ng/mL and/or hormonal replacement with testosterone.

Exocrine gonadal dysfunction was observed in 18 of 49 patients (36.7%), varying from 21.7% (n = 5/23) in nonexposed patients to 50% (n = 13/26) in patients exposed to oral cyclophosphamide. In the exposed patients (n = 26), nine of 20 received ≤4.5 g/m² and four of six received ≥4.5 g/m² of oral cyclophosphamide. The median cumulative dose of oral cyclophosphamide of patients without exocrine gonadal dysfunction was 0 g/m² (range, 0–9) versus 3.1 g/m² (range, 0–9) in cases of exocrine gonadal dysfunction (Figure‐2). Three patients had sperm count analysis; two of them had azoospermia, including one with prostate primary treated with surgery and external beam radiation therapy without abnormal hormonal dosages (details in Table S1). Two patients suffered from endocrine gonadal dysfunction and were treated with hormonal replacement therapy: one with post‐RT hypopituitarism and one with combined endocrine and exocrine dysfunction after orchiectomy and external RT for a paratesticular RMS.

FIGURE 2.

Dot plot and box plot representations of cumulative dose of oral cyclophosphamide according to exocrine gonadal status. The bottom and top edges of the box indicate the interquartile range. The line inside the box indicates the median value.

In univariate analysis, exposure to oral CPM (OR, 3.6; 95% CI, 1.03–12.61, p = .045) and age at diagnosis (compared to the reference category 0–5 years: OR_{5–10 years}, 6.14; 95% CI, 1.30–34.21 and OR_{>10 years}, 10.50; 95% CI, 1.62–68.07, p = .041) were significantly associated with exocrine gonadal dysfunction. The highest cumulative dose of oral CPM (>4.5 g/m²) tended to be more toxic (compared to the reference category of nonexposed patients: OR_≤4.5 g_/m², 2.95; 95% CI, 0.78–11.09 and OR_>4.5 g_/m², 7.20; 95% CI, 1.01–51.39, p = .094) but this dose–effect relationship was difficult to analyze because only six patients were exposed to a cumulative dose >4.5 g/m². Other potential risk factors (primary tumor site, pubertal status at diagnosis, orchiectomy, pituitary gland irradiation, and testicular irradiation) were not significantly associated with exocrine gonadal dysfunction (Table 4). In multivariable analysis, age at diagnosis and exposure to maintenance treatment remained significantly associated with exocrine gonadal dysfunction.

TABLE 4.

Risk factor analysis of male gonadal toxicity using univariate and multivariable logistic regression models (n = 49).

Patients’ characteristics	Univariate analysis				Multivariable analysis
	n/N (%)	OR	95% CI	p	OR	95% CI	p
Age at diagnosis (years)
0–5 years	2/17 (11.8)	Ref		.04	Ref		.03
5–10 years	9/19 (47.4)	6.14	1.10–34.21		9.61	1.49–62.15
>10 years	7/13 (53.8)	10.50	1.62–68.07		14.10	1.89–105.33
Primary tumor
Other	10/31(32.3)	Ref		.39
Genito‐urinary tumor	8/18 (44.4)	1.68	0.91–3.09
Pubertal status at diagnosis, No. (%)
Pre‐pubertal	12/36 (33.3)	Ref
Intra‐ or post‐pubertal	6/13 (46.2)	1.71	0.47–6.24	.41
Oral CPM, No. (%)
No	5/23 (21.7)	Ref			Ref		.02
Yes	13/26 (50)	3.6	1.03–12.61	.04	5.45	1.30–22.92
No	5/23 (21.7)	Ref		.09
≤4.5 g/m2	9/20 (45.0)	2.95	0.78–11.09
>4.5 g/m2	4/6 (66.7)	7.20	1.01–51.39
Intravenous CPM, No. (%)
No	15/43 (34.9)	Ref
Yes	3/6 (50)	1.87	0.33–10.41	.48
Orchiectomy, No. (%)
No	16/45 (35.6)	Ref
Yes	2/4 (50.0)	1.81	0.23–14.12	.57
Pituitary gland radiation therapy, No. (%)
0–24 Gy	15/28 (53.5)	Ref
>24 Gy	6/21 (28.6)	0.53	0.16–1.78	.31
Testicular radiation therapy, No. (%)
0–1 Gy	12/34 (35.2)	Ref
>1 Gy	6/15 (40.0)	1.22	0.35–4.27	.75

Note: Variables significantly associated with gonadal toxicity in univariate analysis at p value threshold 0.05 were included in multivariable analysis. n = number of patients with events (gonadal toxicity), N = number of evaluated patients.

Abbreviations: CI, confidence interval; CPM, cyclophosphamide; OR, odds ratio.

DISCUSSION

In this analysis of 86 5‐year‐old RMS male survivors with an attained age of at least 12 years treated in HR/VHR groups in RMS2005 trial in France, exocrine gonadal dysfunction was reported in 37% of explored patients. Exposure to oral cyclophosphamide as a maintenance treatment and older age at diagnosis were significant and independent risk factors of exocrine gonadal dysfunction (p = .021 and .025, respectively).

The gonadal toxicity of alkylating agents is well‐known, especially with conventional discontinuous intravenous cyclophosphamide at a cumulative dose >6 g/m² in males. ⁶ , ⁷ , ⁸ , ⁹ , ¹⁰ However, gonadal effect of low‐dose continuous oral cyclophosphamide is less well known, especially in addition to other alkylating agents, like ifosfamide, as used in nonmetastatic HR/VHR RMS. ³

In our analysis, patients not exposed to oral cyclophosphamide but to ifosfamide at a median cumulative dose of 54 g/m² (corresponding to nine IVA courses) had a low but not null probability of gonadal dysfunction (22%). In previous studies, a CED was used to describe gonadal toxicity according to the type of alkylating agents used. ¹² For ifosfamide, the CED is 1:4. However, this ratio was estimated based on equivalent hematotoxicity and not on equi‐gonadotoxicity. The gonadal toxicity we observed in patients exposed to ifosfamide 54 g/m² was less than what we could have expected with cyclophosphamide 13.5 g/m², which is associated with a very high risk of gonadal damage. The difference between these two alkylating agents in terms of gonadal toxicity was previously highlighted. ⁶ , ⁹ Consequently, based on these results, we think that the use of CED in clinics to predict the risk of gonadal toxicity in patients exposed to ifosfamide‐only should be reconsidered.

When ifosfamide was combined with oral cyclophosphamide in maintenance, gonadal toxicity was up to 50% higher, highlighting the additive toxicity of both alkylating agents. However, the specific gonadal toxicity of oral cyclophosphamide in males is poorly described. In childhood cancers, cyclophosphamide is mainly used as intravenous injections contrary to rheumatic diseases such as lupus and nephrotic syndromes, in which very few studies report gonadal function after oral cyclophosphamide exposure. ¹⁵ , ¹⁶ Consequently, it is difficult to conclude to the comparative dose–effect relationship between oral and intravenous cyclophosphamide, and the question of the effect of the cumulative dose of oral cyclophosphamide remained. In our study, patients were exposed to various doses of oral cyclophosphamide (from 0.7 to 9 g/m²). This variability can be explained by reductions of dose in patients with hematological toxicity and protocol deviations with the pursue of maintenance treatment up to 12 cycles in patients with VHR RMS. Only a few patients were exposed to oral cyclophosphamide at a cumulative dose of >4.5 g/m², resulting in underpowered analysis of the dose–effect relationship. This question will be very important to study in the FAR‐RMS multicentric randomized trial (ClinicalTrials.gov identifier: NCT04625907) currently comparing 6 (total oral cyclophosphamide cumulative dose: 4.2 g/m²) versus 12 months (8.4 g/m²) and 12 versus 24 (16.8 g/m²) months of maintenance treatment with oral cyclophosphamide in patients with HR and VHR RMS, respectively.

We also reported a significant deleterious impact of age >5 years on gonadal function. No significant correlation between gonadal dysfunction and age or pubertal status at diagnosis with cyclophosphamide or ifosfamide exposure was reported in previous studies. ⁸ , ¹⁰ , ¹⁷ , ¹⁸

These results are still preliminary, because we had a limited number of included patients and a relatively young population at latest follow‐up. The follow‐up has to be maintained to study a potential change of testicular function with time.

Moreover, the small number of patients may have contributed to underpowered analysis resulting in significant association in univariate analysis only for age at diagnosis and exposure to maintenance treatment that are consequently the only two variables included in the multivariable model. This could be a limitation to the multivariable analysis, and further studies including more patients are needed.

With only 49 of 86 patients with available gonadal function assessment, this study also indicated that clinicians should be more aware of the treatment‐associated gonadal toxicity and it encourages better long‐term gonadal follow‐up.

However, these first results on the gonadal impact of the combination of ifosfamide and oral cyclophosphamide supports the promotion of fertility preservation in patients with HR and VHR RMS receiving maintenance treatment with low‐dose continuous cyclophosphamide. For post‐pubertal patients, a sperm sample cryopreservation should always be proposed before any treatment. For pre‐pubertal patients, testicular tissue banking could be proposed, especially in patients older than 5 years and exposed to maintenance for at least 1 year (corresponding to a cumulative dose of oral cyclophosphamide ≥8.4 g/m²), even if restoration methods are still under development. ⁶ , ¹⁹ , ²⁰ Tumor primary site and testicular irradiation should also be considered in the discussion, even if they were not significantly associated with gonadal dysfunction in our study. ²¹ Unilateral testicular transposition is already proposed to patients receiving brachytherapy for bladder/prostate RMS and to patients with testicular tumors receiving external beam scrotal radiation to reduce the dose on the transposed testis and consequently the risk of radiation‐induced testicular damage. ²²

In conclusion, in males with nonmetastatic HR and VHR RMS, maintenance treatment with oral cyclophosphamide seems to induce additional gonadal damage to the ifosfamide‐based chemotherapy regimen used in induction. Consequently, fertility preservation could be considered in patients exposed to maintenance, especially over patients over 5 years old and exposed to higher dose. Further studies (e.g., in a larger international cohort of RMS 2005 survivors) are necessary to determine more accurately the dose–effect relationship of low‐dose continuous cyclophosphamide on testicular function to better specify the recommended fertility preservation practices according to the duration of maintenance treatment. Complementary study on female gonadal toxicity of oral cyclophosphamide is also necessary and should be planned in RMS2005 as well as in FAR RMS. Ifosfamide‐induced renal dysfunction is another potential persistent toxicity in the long term as well as doxorubicin‐related cardiac damage, justifying systematic monitoring of gonadal, renal, and cardiac functions for patients treated for HR and VHR RMS.

AUTHOR CONTRIBUTIONS

Lea Rossillon: Conceptualization; data curation; formal analysis; visualization; writing—original draft; and writing—review and editing. Cécile Thomas‐Teinturier: Visualization; writing—original draft; and writing—review and editing. Daniel Orbach: Visualization; writing—original draft; and writing—review and editing. Marie‐Dominique Tabone: Visualization; writing—original draft; and writing—review and editing. Amandine Bertrand: Visualization; writing—original draft; and writing—review and editing. Sophie Ansoborlo: Visualization; writing—original draft; and writing—review and editing. Anne‐Sophie Defachelles: Visualization; writing—original draft; and writing—review and editing. Angelique Rome: Visualization; writing—original draft; and writing—review and editing. Stéphanie Haouy: Visualization; writing—original draft; and writing—review and editing. Dominique Plantaz: Visualization; writing—original draft; and writing—review and editing. Stéphanie Bolle: Visualization; writing—original draft; and writing—review and editing. Valérie Bernier‐Chastagner: Visualization; writing—original draft; and writing—review and editing. Florent Guerin: Visualization; writing—original draft; and writing—review and editing. Sabine Sarnacki: Visualization; writing—original draft; and writing—review and editing. Pascale Philippe‐Chomette: Visualization; writing—original draft; and writing—review and editing. Rodrigue Allodji: Visualization; writing—original draft; and writing—review and editing. Laura Lenez: Visualization; writing—original draft; and writing—review and editing. Lucy Métayer: Visualization; writing—original draft; and writing—review and editing. Virginie Barraud‐Lange: Visualization; writing—original draft; and writing—review and editing. Véronique Minard‐Colin: Conceptualization; visualization; writing—original draft; and writing—review and editing. Brice Fresneau: Conceptualization; visualization; formal analysis; data curation; writing—original draft; and writing—review and editing. All authors played an important role in interpreting the results; drafted or revised the manuscript; approved the final version; and agreed to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

CONFLICT OF INTEREST STATEMENT

Sophie Ansoborlo reports fees for professional activities from Bordeaux Hospital Center. Véronique Minard‐Colin reports grant and/or contract funding from F. Hofmann‐La Roche. Daniel Orbach reports consulting fees from Bayer Healthcare, EUSA Pharma (US) LLC, F. Hofmann‐La Roche, and Limburg Sterk Merk; fees for data and safety monitoring from Eli Lilly and Company; and grant and/or contract funding from Medscape. Marie‐Dominique Tabone reports fees for professional activities from Assistance Publique‐Hôpitaux de Paris. The other authors declare no conflicts of interest.

Supporting information

Supplementary Material

Rossillon L, Thomas‐Teinturier C, Orbach D, et al. Impact of maintenance treatment on male gonadal function in patients treated for localized rhabdomyosarcoma in RMS2005 trial in France. Cancer. 2025;e35918. doi: 10.1002/cncr.35918

DATA AVAILABILITY

The study data are available on request.