Intermediate Risk Rhabdomyosarcoma in very young children (less than or equal to 24 months) treated on the prior Children’s Oncology Group protocol ARST1431

Abstract

Background

Rhabdomyosarcoma is the most common soft tissue sarcoma in pediatrics and infants are reported to have inferior outcomes. ARST1431 recommended delayed primary excision (DPE) while discouraging radiation (RT) deviations. This analysis compared outcomes among very young and older patients in this study.

Procedures

Demographic and tumor characteristics were summarized using frequencies and percentages. Local failure (LF) frequency was summarized by age group, RT and DPE status. Event-free survival (EFS) and overall survival (OS) were estimated using the Kaplan–Meier method and compared between age groups using the log-rank test.

Results

Of 296 patients, 48 (16.2%) were <24 months: 14 (4.7%) < 12 months, 34 (11.5%) 12–24 months – compared with 248 (83.8%) ≥24 months. DPE was performed in: 7 (50%), <12 months; 9 (26%), 12–24 months; 56 (22%), ≥24 months. RT was given to 9 (64%), 27 (79%), and 195 (79%) of <12, 12–24, and >24 months, respectively. LF in DPE+RT was 8.3% and 6.1%, < 24 months and ≥ 24 months, respectively, compared to 29.3% and 20.5% for RT without DPE. The 4-year EFS did not vary: 56% (95% CI= 26.5, 86.0), 68% (95% CI = 51.2, 84.1), and 61% (95% CI= 54.2, 68.3) <12, 12 −24, and >24 months old, respectively.

Conclusions

Outcomes in younger children are comparable to older children with local failure more common in those without DPE. Optimization of local control will be important for future studies.

Introduction

Rhabdomyosarcoma (RMS) is the most common soft tissue sarcoma in children and accounts for approximately 2.7% of all cancers in individuals under 14 years of age [1]. Despite therapeutic advances, age remains a critical prognostic factor, with infants—particularly those younger than 12 months—consistently demonstrating inferior outcomes compared to older children [2–7]. Several factors contribute to this disparity, including treatment-related morbidity, toxicity, and long-term sequelae associated with surgery, chemotherapy, and radiotherapy (RT). Historically, these concerns have led to reductions in treatment intensity for very young patients, either through protocol allowances or physician-directed modifications. Additionally, young age is a recognized risk factor for chemotherapy-induced hepatopathy, often necessitating dose reductions in this population [8].

Infants have traditionally been excluded from chemotherapy dose-escalation strategies implemented in prior trials [7]. For example, in IRS-IV, IRS-V, and D9803, infants initially received 50% of the prescribed chemotherapy dose, with subsequent adjustments based on tolerance [6]. Furthermore, dosing strategies varied by protocol, utilizing either weight-based or body surface area calculations. These modifications, while intended to mitigate toxicity, may have inadvertently compromised treatment efficacy.

Radiotherapy deviations have also been a persistent challenge. In earlier studies, RT modifications for patients younger than 24 months were considered protocol deviations. Despite recommendations, RT was frequently withheld by local investigators or declined by parents, resulting in major RT deviations in approximately 42% of infants across IRS-IV, IRS-V, and D9803 [6]. These deviations likely contributed to poorer outcomes, including increased local failure rates (30% in infants versus 13% in older patients) and reduced overall survival.

Recognizing these limitations, the Children’s Oncology Group (COG) trial ARST1431 introduced key modifications aimed at optimizing local control while minimizing toxicity. Specifically, RT deviations were permitted only for patients under 24 months of age with FOXO1-negative disease. In contrast to prior studies, ARST1431 also encouraged delayed primary excision (DPE) following initiation of chemotherapy, reflecting a strategic shift toward surgical consolidation after tumor shrinkage. The standardized approach to local therapy in ARST1431 provides a unique opportunity to evaluate how these modifications influenced local failure and overall outcomes in younger children with intermediate-risk RMS.

In this analysis, we sought to compare clinical outcomes between infants and older pediatric patients enrolled in ARST1431, with particular attention to patterns of local therapy, protocol adherence, and survival. Understanding these differences is critical for refining treatment strategies and improving prognosis for this vulnerable population.

Materials & Methods

Study Population

This study included 296 evaluable patients who were treated on ARST1431[1]. Records of patients <12 months and those 12–24 months of age were specifically evaluated for clinical characteristics, event free survival (EFS), overall survival (OS), and patterns of disease recurrence or progression. Intermediate risk in ARST1431 was defined as the following: embryonal RMS Stage I, Group III (non-orbit), Stage 3, Group I/II, Stage 2/3, Group III, Stage 4, Group IV < 10 years old and alveolar RMS Stages 1 – 3, Groups I-III. Consent was signed by the parents / guardians of these children at the time of initial enrollment to this study protocol.

Treatment

Chemotherapy consisted of vincristine, dactinomycin, cyclophosphamide, and irinotecan with a 1:1 randomization to add temsirolimus. Cyclophosphamide dosing was 1200 mg/m² for patients with a BSA of greater than or equal to 0.6 m² and based on a dosing table for patients with a smaller BSA. Local control was surgery, radiation, or both. An attempt at surgical excision required the intention of complete excision of disease with negative margins unless it was considered to lead to unacceptable loss of function or form. DPE at week 9 was encouraged on this protocol.

RT was indicated for all patients except those with Group I, FOXO1 negative or FOXO1 indeterminate disease. On this protocol, RT deviations were allowed at the discretion of treating clinicians for patients ≤ 24 months who were FOXO1 negative. Deviation from protocol for all patients who had FOXO1 positive disease was considered a protocol violation. RT of the primary site was performed at week 13 and to metastatic sites at week 43. Target dose of RT for tumors greater than 5 cm was 59.4 Gy. In smaller tumors, the target dose was 50.4 Gy. Importantly, for patients who had a DPE with a gross total resection, RT could be decreased to 36 Gy with negative margins, and 41.4 Gy with microscopically positive margins.

Statistical Analysis

Patient characteristics were summarized and compared across age groups (<12 months, 12–24 months, and >24 months). Frequencies and percentages were used to describe categorical variables. The distribution of characteristics across age groups was compared using the Chi-square test or Fisher’s exact test, as appropriate.

EFS—defined as the time from enrollment to the first occurrence of secondary malignancy, relapse/progression, or death—was estimated using the Kaplan–Meier method with 95% confidence intervals. Differences in EFS between age groups were evaluated using the log-rank test.

Local failure (LF) rates were estimated within subgroups defined by age, RT, and DPE status. A side-by-side bar chart was used to visualize the LF rates within each subgroup.

Results

Patient & Disease Characteristics

Of the 296 patients, 48 (16.2%) were very young: 14 (4.7%) <12 months, 34 (11.5%) 12–24 months, compared with 248 (83.8%) were ≥24 months (Table 1).

Table 1:

Patient Characteristics

	<12 months (N=14)	12–24 months (N=34)	>24 months (N=248)	Total (N=296)	P-value
Fusion, n (%)					0.59441
FOX01-	11 (78.6%)	29 (85.3%)	181 (73.0%)	221 (74.7%)
FOX01+	3 (21.4%)	5 (14.7%)	64 (25.8%)	72 (24.3%)
Unknown	0 (0.0%)	0 (0.0%)	3 (1.2%)	3 (1.0%)
Sex, n (%)					0.12732
Male	8 (57.1%)	26 (76.5%)	145 (58.5%)	179 (60.5%)
Female	6 (42.9%)	8 (23.5%)	103 (41.5%)	117 (39.5%)
Primary tumor sites, n (%)					0.03111
Extremity	5 (35.7%)	6 (17.6%)	35 (14.1%)	46 (15.5%)
GU B/P	2 (14.3%)	9 (26.5%)	24 (9.7%)	35 (11.8%)
GU, non-B/P	1 (7.1%)	1 (2.9%)	10 (4.0%)	12 (4.1%)
H&N	2 (14.3%)	4 (11.8%)	42 (16.9%)	48 (16.2%)
Orbit	0 (0.0%)	0 (0.0%)	3 (1.2%)	3 (1.0%)
Other, unspecified	0 (0.0%)	4 (11.8%)	13 (5.2%)	17 (5.7%)
PM	2 (14.3%)	2 (5.9%)	74 (29.8%)	78 (26.4%)
Retroperineum	2 (14.3%)	5 (14.7%)	33 (13.3%)	40 (13.5%)
Trunk	0 (0.0%)	3 (8.8%)	14 (5.6%)	17 (5.7%)
IRS group, n (%)					0.11201
I	0 (0.0%)	2 (5.9%)	9 (3.6%)	11 (3.7%)
II	4 (28.6%)	1 (2.9%)	13 (5.2%)	18 (6.1%)
III	9 (64.3%)	27 (79.4%)	198 (79.8%)	234 (79.1%)
IV	1 (7.1%)	4 (11.8%)	28 (11.3%)	33 (11.1%)
T Stage, n (%)					0.88841
1	3 (21.4%)	5 (14.7%)	43 (17.3%)	51 (17.2%)
2	2 (14.3%)	8 (23.5%)	36 (14.5%)	46 (15.5%)
3	8 (57.1%)	17 (50.0%)	141 (56.9%)	166 (56.1%)
4	1 (7.1%)	4 (11.8%)	28 (11.3%)	33 (11.1%)
Tumor size, n (%)					0.64832
<5	6 (42.9%)	14 (41.2%)	86 (34.7%)	106 (35.8%)
>=5	8 (57.1%)	20 (58.8%)	162 (65.3%)	190 (64.2%)
Nodal Clinical, n (%)					0.10041
No, N0	8 (57.1%)	27 (79.4%)	146 (58.9%)	181 (61.1%)
Yes, N1	6 (42.9%)	6 (17.6%)	98 (39.5%)	110 (37.2%)
Not Evaluated/Unknown	0 (0.0%)	1 (2.9%)	4 (1.6%)	5 (1.7%)
Nodal Path, n (%)					0.01071
No, N0	10 (71.4%)	32 (94.1%)	208 (83.9%)	250 (84.5%)
Yes, N1	4 (28.6%)	1 (2.9%)	40 (16.1%)	45 (15.2%)
Not Evaluated/Unknown	0 (0.0%)	1 (2.9%)	0 (0.0%)	1 (0.3%)

¹Fisher’s Exact Test p-value;

²Chi-Square p-value.

Chemotherapy

Chemotherapy was completed in 12 (86%), 22 (65%), and 152 (61%) of those who were <12, 12–24, >24 months, respectively. 0 (0%), 5 (23%), and 56 (37%) of patients were removed from protocol therapy for reasons other than disease progression/withdrawal of consent for <12, 12 −24, and >24 months old, respectively.

Local Therapy - DPE

Of the 48 patients <24 months old, DPE was performed on 7 (50%) of those <12 months, 9 (26%) of 12–24 months and 56 (22%) of ≥24 months. For those who had a DPE, margin status (for data available) was as follows: < 12 months: 2/2 positive, 12 – 24 months: 2/5 positive and 3/5 negative. Only one patient, who was 14 months old, was considered to have a resection of a vital structure that included the small bowel and bladder.

Local Therapy - Radiation

RT was given to 9 (64%), 27 (79%), and 195 (79%) of < 12, 12 −24, and >24 months old, respectively. RT without minor or major deviation, as prescribed by protocol, for those who had complete data was given to, 8 (89%) 17 (63%), and 153 (78%) of <12, 12 −24, and >24 months old, respectively.

Reasons for not receiving RT are listed in Appendix A.

Outcomes

The four-year EFS across all three cohorts was around 62% (95% CI= 55.4, 68.1). EFS and OS did not vary by age group. EFS was 56.3% (95% CI = 26.5%, 86.0%), 67.6% (95% CI = 51.2%, 84.1%), and 61.2% (95% CI = 54.2%, 68.3%) for patients < 12 months, 12 – 24 months and ≥ 24 months. OS was 84.6% (95% CI = 61.6%, 100.0%), 79.4% (95% CI = 65.3%, 93.5%), and 73.2% (95% CI = 66.8%, 79.6%) (Figure 1). There was no statistically significant difference in the 4 year EFS or OS for patients who had protocol deviations for dose reductions in either cyclophosphamide or RT.

Figure 1:

Kaplan Meier Curves

LF alone occurred in 3 (21%), 5 (15%), and 37 (15%) of < 12, 12 – 24, and ≥24 months patients. Distant failure alone occurred in 1 (7%), 2 (6%), and 37 (15%) of patients < 12, 12 – 24, and ≥24 months. In those respective age groups, distant failure + LF occurred in 0 (0%), 4 (12%), and 9 (4%). Of 12 local failures in patients <24 months, only 1 had a DPE.

LF in those <24 months who had RT and DPE was 8.3% compared with 29.2% in those without DPE. (Figure 2). Characteristics of LF for those patients <24 months can be found in table 2.

Figure 2.

Impact of DPE on Local Failure Rate by Age and RT Status

Table 2.

Characteristics for those with local failure

Age (mo)	Sex	Disease Site	Histology	FOXO1 neg/pos	Stage	Group	Max Tumor Size (cm)	RT yes/no	DPE yes/no	Completed all protocol chemo yes/no	time to LF from start of chemo (days)	RT should have had it per the protocol	Progressed while on protocol therapy	PtStatus
0.03	Female	Bladder	ERMS	Negative	3	III	7.6	Yes	Yes	Yes	586	RT	No	Alive
1.22	Male	Forearm	ARMS	Positive	3	II	8.0	Yes	No	Yes	664	RT	No	Died
5.62	Female	Thigh	ERMS	Negative	3	III	5.5	No	No	No	18	No RT	Yes	Alive
11.86	Male	Buttock	Mixed RMS	Positive	3	II	5.3	Yes	No	Yes	1131	RT	No	Alive
15.97	Female	Arm	ARMS	Negative	3	III	6.6	No	No	No	101	No RT	Yes	Died
16.30	Male	Perineum	ARMS	Positive	3	III	6.6	Yes	No	Yes	2119	RT	No	Alive
17.02	Female	Parotid	NOS	Negative	1	III	5.6	Yes	No	No	188	RT	Yes	Died
17.91	Male	Perineum	ERMS	Negative	4	IV	5.5	Yes	No	No	131	RT	Yes	Died
19.78	Female	Retroperitoneum	Spindle cell	Negative	3	III	14.4	Yes	No	No	393	RT	Yes	Died
21.16	Male	Oral Cavity	ERMS	Negative	1	III	8.4	No	No	No	103	No RT	Yes	Died
21.82	Male	Parotid & PM Extension	ERMS	Negative	3	III	6.5	Yes	No	No	911	RT	No	Alive
22.97	Male	Prostate	ERMS	Negative	4	IV	14.5	No	No	No	64	No RT	Yes	Died
23.16	Male	Pelvis, Site Indeterminate	ERMS	Negative	3	III	6.5	No	No	No	64	No RT	Yes	Alive

Discussion

Herein, we demonstrate, for the first time, that patients <24 months, have similar outcomes to older patients with intermediate risk RMS. One of the factors that may have contributed to this equalization in outcome is the utilization of DPE combined with RT thereby decreasing local failures.

The proportion of children less than 1 year enrolled in ARST1431 was 3% in D9803 and 7% in ARST0531. 12% of patients in D9803 and 10% of patients in ARST0531 were between 12 and 24 months [2, 3]. In earlier trials, the total cumulative dose of alkylating agents was 25.1– 30.8 g/m² in D9803 and 8.4 – 16.8 g/m² in ARST0531, compared with 12.6 g/m² in ARST1431[4].

The proportion of patients receiving radiation therapy (RT) was comparable across all three studies; however, the rate of delayed primary excision (DPE) was low. Notably, very young children demonstrated inferior event-free survival (EFS) compared to older children in both prior studies. Melampati et al., reported a significantly poorer outcome in 76 (4.1%) children <12 months compared to older children with a 5-year failure-free survival of 57% (44%, 67%) compared to 81% (79%, 83%) for ages 1–9 years. Twenty-three of 32 infants with treatment failure had local recurrence/ progression. DPE was performed in some capacity in 14 patients, but, it remains unclear what the denominator was. Importantly, when compared to Melampati et al, compliance with RT target doses in our study was significantly improved (88% compared to 42%) [5].

In contrast, with chemotherapy doses more aligned with ARST1431, infants less than 12 months enrolled on the EpSSG RMS 2005 study had similar EFS and OS when compared to older children[6].

The benefit of DPE in reducing LF has been previously documented [7]. Despite radiation being delivered at prescribed doses, most treatment failures remain local. Among patients who received RT as assigned, 16% still experienced local failure, underscoring the need for improved strategies to enhance local control. Notably, the local control review for the entire ARST1431 trial confirmed that DPE was associated with lower LF rate and there was no added benefit of RT dose escalation for tumors larger than 5 cm [8]. Tumors that are smaller and/or in more favorable locations are typically more amenable to DPE, resulting in confounding factors when directly comparing the outcomes with RT alone to RT + DPE. In this study 35% of patients less than 12 months had extremity tumors compared to 17% of patients 12 – 24 months and 14% of patients older than 24 months and, generally speaking, extremity tumors are far more amenable to DPE than other tumor sites.

Despite the implementation of DPE and RT, local failure continues to be dominant problem in very young children. We were able to demonstrate the lack of difference in EFS between younger and older children, but at this time, we can only hypothesize that this may be due to improvements in imaging, RT planning, quality of DPE and overall consistency in delivery of local therapy to younger patients. We remain limited in truly understanding the long-term effects of various RT dosing regimens with and without DPE in long term survivors.

In summary, with current treatment protocols, younger children fare similar to older children and ongoing attention to local therapy remains imperative in future studies.

Abbreviation

COG

Children’s Oncology Group

DPE

Delayed Primary Excision

EFS

Event Free Survival

LF

Local Failure

OS

Overall Survival

RMS

Rhabdomyosarcoma

RT

Radiation Therapy