Soft Tissue Sarcoma or Malignant Mesenchymal Tumors in the First Year of Life: Experience of the International Society of Pediatric Oncology
http://www.100md.com
《临床肿瘤学》
the Institut Curie, Paris
Institut Gustave Roussy, Villejuif, France
Hospital Val d’Hebron, Barcelona, Spain
Jeroen Bosch Ziekenhuis, ’s-Hertogenbosch, the Netherlands
Royal Hospital for Children, Bristol, United Kingdom
ABSTRACT
PURPOSE: To describe the outcome of infants with a histologically confirmed diagnosis of malignant mesenchymal tumor (MMT) included in the International Society of Paediatric Oncology studies MMT 84 and MMT 89.
PATIENTS AND METHODS: One hundred two infants ( 12 months old) were included. Twenty-four children were less than 3 months old, and 16 were less than 1 month old. Sixty-four patients had rhabdomyosarcoma (RMS), 26 had undifferentiated sarcoma, and 12 had other histology. Clinical TNM stage was stage I (41%), II (39%), III (6%), and IV (14%). First-line treatment was ifosfamide, vincristine, dactinomycin, whereas the second-line combination consisted of either cisplatin and doxorubicin (in MMT 84) or vincristine, carboplatin, etoposide/teniposide (in MMT 89). Chemotherapy doses were adapted to age. Local therapy was conservative surgery as often as possible.
RESULTS: After a median follow-up of 7.8 years (range, 0.1 to 13 years), 5-year overall survival (OS) and event-free survival rates were 66% and 55% for the total study population and 72% and 60% for nonmetastatic patients, respectively. Only two of 13 stage IV patients survived. Sixty-seven percent of newborn infants survived. Infants with alveolar subtype had a poorer survival than those with non-RMS MMT or nonalveolar RMS (5-year OS, 37% v 75% or 82%, respectively; P = .002). When compared with older children with MMT, young age does not seem to be an important prognostic factor.
CONCLUSION: OS was satisfactory even when local treatment was not aggressive, although the prognosis was poor for infants with alveolar RMS or metastatic tumors. Chemotherapy toxicity was manageable with appropriate dose modification.
INTRODUCTION
More than 75% of soft tissue tumors diagnosed between birth and 12 months of age are pathologically benign1 and include diagnoses such as infantile hemangioendothelioma, lymphangioma, or infantile myofibromatosis. Ten percent of tumors are borderline lesions, and only 15% of such tumors in this age group are malignant. Previous reports suggest that the main specific diagnoses of malignant mesenchymal tumors (MMTs) are rhabdomyosarcoma (RMS) and peripheral primitive neuroectodermal tumor,1 although a significant proportion of these infantile tumors are undifferentiated.2-5 Overall, 5% to 11% of childhood MMTs are reported in infants before the age of 1 year.6-9 Prognosis depends on histology, although, for the same diagnosis, the prognosis may be more favorable in infants than in older children. For example, survival rates as high as 100% have been reported in congenital fibrosarcoma.10,11 However, the outlook for neonatal RMS has been reported to be less favorable than for older children.12 In general, age seems to be a prognostic factor in RMS.3,6-8 Recent studies have shown that infants with RMS have a less satisfactory outcome than older children.8,9,13 Clinical management remains difficult in infants because of the immaturity of many physiologic systems.14 The risk of long-term sequelae, particularly after radical surgery, radiotherapy, or the use of alkylating agent chemotherapy, remains a constant preoccupation in the choice of treatment2 at this age. We report the characteristics and outcome of a group of 102 infants less than 13 months old with MMTs registered onto the International Society of Pediatric Oncology (SIOP) MMT 8415 and MMT 8916 protocols, and we compare these results with the results observed in older children treated in the same studies.
PATIENTS AND METHODS
Inclusion Criteria
Infants were prospectively registered onto the MMT 84 and MMT 89 studies from 1984 to 1995. Diagnoses were confirmed by the SIOP Pathology Review Committee. The MMT 84 protocol registered children diagnosed between the ages of 3 months and 18 years, whereas the MMT 89 protocol included children from birth to 18 years. Children aged more than 3 months who were registered in MMT 89 and who had metastatic disease were treated according to a separate European Intergroup study.17 According to European law at the time of the studies, no informed consent was required for enrollment onto these studies. Data on a total of 102 infants (20 registered onto MMT 84 and 82 registered onto MMT 84) were analyzed for this report.
Exclusion Criteria
All children with a reviewed diagnosis indicating a benign tumor (fibromatosis, inflammatory tumor, hamartoma, or myxoma/myxofibroma) or children with tumors that could not be confirmed histologically were excluded from this study. Primary tumor site was classified into the following six standard subgroups: nonparameningeal orbit; parameningeal head and neck (including orbit with parameningeal extension); nonparameningeal head and neck; genitourinary, including bladder/prostate or nonbladder/prostate (paratesticular or vaginal/uterine sites); limbs; and other sites.18
Clinical staging was defined according to the SIOP–International Union Against Cancer TNM system.19 Postsurgical staging was defined in the protocol by the SIOP pathologic TNM staging system, but these data were also presented according to the Intergroup Rhabdomyosarcoma Study group.
Histologic Classification
Since 1984, all tumors have been reviewed by an international panel of pathologists and classified as RMS (according to the new RMS international classification20) or as non-RMS MMT. The diagnostic classification did not take into account any data from cytogenetic or molecular genetic studies on individual tumors. Undifferentiated sarcomas composed of spindle cells were considered separately from the other undifferentiated tumors and were assessed together with fibrosarcomas because of their similar behavior.
Treatment
The details of the MMT 84 and MMT 89 protocols have been published previously.15,16 The strategy used for stage I and II disease was determined by the extent of primary surgery. Patients with complete primary resection (group 1) received chemotherapy consisting of three courses of ifosfamide, vincristine, and dactinomycin (IVA) in MMT 84 but received only four courses of vincristine and dactinomycin (VA) in MMT 89 (Table 1). Incomplete primary resection (groups 2 and 3) was followed by six to 10 courses of IVA (ifosfamide 6 g/m2/course) in MMT 84 and by six courses of IVA in MMT 89 (ifosfamide 9 g/m2/course). In cases of partial response or progressive disease, second-line chemotherapy consisted of cisplatin and doxorubicin in MMT 84 and vincristine, carboplatin, and teniposide (or etoposide) in MMT 89. To minimize late effects, further local therapy was administered only to patients who did not achieve complete remission (CR) with primary surgery and/or subsequent chemotherapy. Whenever possible, this was limited to conservative surgery, and radiotherapy was avoided in these young patients. No specific guidelines concerning radiotherapy (dose or field) were specified apart from the use of brachytherapy whenever possible. Significant local therapy was defined as either large (major surgery but with no predictable major long-term functional or cosmetic consequences and/or external-beam radiotherapy) or radical (major surgery with predictable functional or cosmetic consequences; eg, total cystectomy, orbital exenteration, and amputation).
All patients with either metastatic, node-positive or parameningeal tumors registered onto the MMT 89 protocol were treated with an intensified six-drug chemotherapy regimen, which included IVA; carboplatin, epirubicin, and vincristine; and ifosfamide, vincristine, and etoposide (Table 1).
In both studies, chemotherapy doses were reduced by 50% in infants under the age of 6 months and by 33% in infants between the ages of 6 and 12 months. When the first course was well tolerated, chemotherapy was increased to full dose for the following courses. Infants less than 1 month old initially received alternative single-agent therapy with VA at 50% of the standard dose.
Toxicity
Chemotherapy toxicity was classified into five grades (grade 0 = no toxicity; grade 4 = life-threatening toxicity) according to the National Cancer Institute Common Toxicity Criteria (version 1) and a specific grading system for nephrotoxicity.21 Toxicity data were systematically and prospectively assessed in children included in the MMT 89 protocol. Long-term complications (local or systemic) were not systematically recorded.
Statistical Methods
Outcome was defined by overall survival (OS) and event-free survival (EFS). For EFS, events were defined as relapse after CR or death from any cause (including progression of disease without achieving CR). If a patient failed to achieve CR, the time-point for failure was the date of death (date of progression was not recorded). This method of calculation, in children who died without achieving remission, does not alter the 5-year EFS rate but may slightly alter the slope of the survival curve over the first few years.
Local control was defined as control of the primary tumor site with disappearance of all clinical and radiologic signs of disease or stable residual radiographic images for 6 months after completion of treatment. Statistical analysis was performed using a general database management system. Survival curves were calculated by the Kaplan-Meier method. The statistical significance of each variable was first tested by the log-rank test (univariate procedure). Some data were then compared with the data of older RMS or non-RMS children (aged 1 to 18 years) included and treated in the SIOP MMT 84 and 89 protocols during the same period.
RESULTS
Patient Population
One hundred two infants under the age of 13 months, who were treated between 1984 and 1995, were analyzed in this study. Most of these infants were registered after 1990 (20 patients in the MMT 84 protocol and 82 in the MMT 89 protocol). This group represented 8% of all children (102 of 1,228 children) entered onto the two studies. Median age at diagnosis was 5 months (range, 0 to 12 months); 16 infants were less than 1 month old, and 24 were less than 3 months old. The sex ratio was 1.9 (67 males and 35 females). Median follow-up time of the whole study population was 7.8 years (range, 0.1 to 13 years).
Histologic Classification
Diagnoses as determined by the pathology review committee are listed in Table 2. RMS tumors were more frequent than non-RMS tumors (64 v 38 tumors). The alveolar subtype was detected in 31% of infants with RMS. Undifferentiated sarcoma was the most frequent non-RMS histologic type.
Primary tumor sites according to histology are listed in Table 3. The primary tumor involved the limbs in 30 patients, corresponding to 20% of embryonal RMS, 30% of alveolar RMS, and 39% of non-RMS. Other sites, such as pelvic, abdominal, and retroperitoneal sites, were frequent and mostly concerned non-RMS MMT. Most of the genitourinary tumors were embryonal RMS. Only four parameningeal tumors and four orbital tumors were observed, but three of the four orbital tumors showed alveolar histology.
Clinical Staging
As shown in Table 4, most tumors were initially localized (stage I or II), although 14% of patients presented with distant metastatic disease at diagnosis. Primary surgical resection was complete in only 13% of patients and resulted in microscopic residual disease in 23%. The majority of patients had macroscopic residual disease after the initial surgical resection or biopsy. An unexpectedly high proportion of children (12%) presented with associated pleural or peritoneal effusion (stage pT3c).
CR was obtained in 77% of the infants (76 of 98 infants) for whom this information was available. The CR rate varied according to stage of disease; the CR rate was 95% in stage I patients (39 of 41 patients), 74% in stage II patients (29 of 39 patients), 57% in stage III patients (four of seven patients), and 36% in stage IV patients (four of 11 patients). Data were missing for four patients. Only 13% of patients achieved CR with initial surgery alone (58% after incomplete initial surgery plus chemotherapy, 21% after chemotherapy followed by secondary surgery, and 7% after chemotherapy ± surgery plus radiotherapy; 3% received brachytherapy). One infant with stage IV disease achieved CR after chemotherapy, radiotherapy, surgery, and high-dose chemotherapy. The treatment received to achieve CR was not reported in one child. Although radiotherapy was not recommended regardless of age, five infants received this modality as first-line treatment based on the clinician’s preference. These infants had a median age of 8 months (range, 5 to 12 months) and had stage II disease (embryonal RMS, n = 4; and embryonal sarcoma, n = 1). Treatment was delivered either by external-beam radiotherapy (44 to 45 Gy; n = 3) to the initial tumor volume with margins (n = 2) or to the residual tumor volume after chemotherapy (n = 1) or by brachytherapy (50 to 65 Gy; n = 2). None of these infants developed local relapse.
Chemotherapy Toxicity
Detailed toxicity data were available only for patients on the MMT 89 protocol. Table 5 lists the percentages of children experiencing grade 3 or 4 toxicity after chemotherapy for whom complete data were available. Toxicity was mainly hematologic and less commonly infectious. This toxicity led to an additional dose reduction in 19% of infants receiving IVA; 33% of infants receiving vincristine, carboplatin, and teniposide (or etoposide); 43% of infants receiving carboplatin, epirubicin, and vincristine; and 60% of infants receiving ifosfamide, vincristine, and etoposide; no reductions were necessary after VA. Five infants who received ifosfamide experienced severe acute renal toxicity (glomerular, n = 1; or tubular, n = 4). These infants were 1, 2, 7, 9, and 12 months old at the time of this toxicity.
Relapse
Twenty-six (34%) of the 76 infants who achieved CR subsequently relapsed a median of 13 months (range, 3 to 40 months) after the start of treatment. Relapse was mainly local (69%), but relapse was metastatic in 23% of patients and local and metastatic in 8% of patients. Further treatment consisted of chemotherapy alone (n = 3), surgery alone (n = 1), chemotherapy and surgery (n = 4), chemotherapy with radiotherapy ± surgery (n = 9), or surgery and radiotherapy (n = 2). No treatment details were available for seven patients. At final analysis, 12 (46%) of 26 relapsing patients with embryonal RMS (n = 4), alveolar RMS (n = 2), undifferentiated spindle-cell sarcoma (n = 1), other undifferentiated sarcomas (n = 3), or non-RMS sarcomas (n = 2) were alive with a median follow-up of 7 years (range, 1 to 10 years) after relapse.
Cause of Death
Thirty-five infants died, mainly from their disease (n = 33). One infant died from toxicity (a 5-month-old infant in whom the dosage of dactinomycin was not correctly reduced according to age and who developed liver failure). One child, who was treated 33 months previously for alveolar RMS, died after treatment for a second malignancy (acute myeloid leukemia).
Survival
The outcome of the whole group showed a 5-year OS rate of 66% and a 5-year EFS rate of 55% (Fig 1). More favorable results were observed in localized tumors (stages I to III: OS, 73% and EFS, 60%; Fig 2) than in metastatic disease (OS and EFS, 17%; 95% CI, 5% to 45%). Survival was better for infants with localized nonalveolar RMS or non-RMS MMT than for infants with localized alveolar RMS (5-year OS, 82%, 75%, and 40%, respectively; P = .004; Fig 3). Similar results were observed for EFS (Fig 4).
Newborns
Sixteen infants were less than 1 month old at diagnosis (10 males and six females). The histologic distribution was as follows: five RMS (one alveolar), two undifferentiated spindle-cell sarcomas, one fibrosarcoma, one peripheral primitive neuroectodermal tumor, one embryonal sarcoma, and six other non-RMS MMT. The primary site was the limbs in eight patients; others sites included nonparameningeal head and neck (n = 3), genitourinary (n = 1), and other sites (n = 4). Clinical stages were as follows: stage I (n = 5), II (n = 9), and IV (n = 2). CR was obtained in nine (64%) of the 14 assessable patients and was achieved with chemotherapy alone in six patients, chemotherapy and surgery in one patient, and surgery alone in two patients (missing data for two patients). Eleven of the 16 infants are alive, with a follow-up ranging from 1.8 to 10 years, and the 5-year OS rate was 67%. Three of the five infants with RMS are alive.
Patients With Metastatic Disease
Thirteen infants (six males and seven females) had metastatic disease at the time of diagnosis (alveolar RMS, n = 5; nonalveolar RMS, n = 3; and undifferentiated sarcomas, n = 5). The primary sites were the limbs (n = 3), parameningeal head and neck (n = 2), paratesticular/vaginal (n = 2), orbit (n = 1), and other sites (n = 5). Metastases were usually multiple and involved bone (n = 3), lungs (n = 3), bone marrow (n = 2), liver (n = 2), CNS (n = 1), or other sites (n = 6). Only three infants achieved CR (with chemotherapy; chemotherapy plus surgery; and chemotherapy, surgery, radiotherapy, and high-dose chemotherapy). Ten of the 12 assessable infants died (one was lost to follow-up) after a median follow-up of 11 months. Only two infants, one with embryonal RMS and one with undifferentiated sarcoma and initial bone and liver metastases, are still alive 81 and 85 months after diagnosis, respectively.
Comparison Between Histologic Types
Fibrosarcoma and undifferentiated spindle-cell sarcoma represented the largest specified diagnostic group among non-RMS MMT. Twelve infants (eight males and four females) had fibrosarcoma (n = 5) or undifferentiated sarcomas composed of spindle cells (n = 7). The age at diagnosis ranged from 7 days to 4 months. The primary site was usually the limbs (n = 7), with other sites in five infants. The initial staging was as follows: stage I (n = 6), II (n = 4), III (n = 1), and unknown (n = 1). When known, CR was obtained in eight of nine patients with chemotherapy (five patients: six courses of IVA; one patient: three courses of IVA; and one patient: three courses of IVA and three courses of cisplatin and doxorubicin) and/or conservative surgery (after chemotherapy in two patients or alone in one patient). In two patients, the residual mass progressively disappeared with time after chemotherapy or surgery alone (missing data for one patient). OS and 5-year EFS rates were 92% (95% CI, 65% to 99%) and 83% (95% CI, 55% to 95%), respectively.
Undifferentiated sarcoma with no pattern of differentiation (other undifferentiated sarcoma) was observed in 19 infants (13 males and six females). Six infants were less than 1 month old at diagnosis, two were aged between 1 and 3 months, two were aged between 3 and 6 months, and nine were older than 6 months. The primary site was mostly other sites (n = 8) or limbs (n = 5). Initial staging was stage I (n = 5), II (n = 8), and IV (n = 5), with missing data for one patient. CR was obtained in nine of 12 patients, when data were available, with surgery only (n = 1), chemotherapy (n = 5), or chemotherapy followed by surgery (n = 3). At 5 years, the OS and EFS rates were 56% (95% CI, 35% to 46%) and 45% (95% CI, 25% to 67%).
The median age of the 64 infants (42 males and 22 females) with RMS was 7 months. The primary sites were as follows: limbs (n = 15), genitourinary nonbladder/prostate (n = 13), genitourinary bladder/prostate (n = 12), head and neck nonparameningeal (n = 8), head and neck parameningeal (n =2), orbit (n = 4), and other sites (n = 10). CR was achieved in 87% of patients. Initial first CR for infants with localized RMS was obtained with surgery only (14%), chemotherapy (50%), chemotherapy followed by conservative surgery (27%), chemotherapy and brachytherapy (3%), chemotherapy followed by surgery and brachytherapy (3%), or chemotherapy, surgery, and external radiotherapy (3%). The EFS and OS rates according to histologic subtype in localized tumors are indicated in Figures 3 and 4. Among the 31 infants alive with a minimum follow-up of 2 years who were treated in the MMT 89 protocol for localized RMS, four were treated by chemotherapy alone, 20 were treated by chemotherapy and conservative surgery, and seven were treated by chemotherapy with significant local therapy. Only two patients were treated with local therapy defined as large, and five were treated with local therapy defined as radical. In summary, 78% of survivors were treated using a local conservative approach.
Comparison With Older Patients
The characteristics and outcome of infants with RMS were compared with those of a group of older children with RMS (age range, 1 to 18 years) who were treated in the same studies. Sex ratio, clinical stage, and CR rate were not significantly different. There was a higher proportion of alveolar RMS subtypes in the infant group than in the older children (31% v 19%, respectively; P = .01). Microscopically incomplete initial surgery was more frequent in infants than in older children (31% v 15%, respectively; P < .01). Primary site was more often genitourinary bladder/prostate and limbs in infants (19% v 9% in older children; and 23% v 15% in older children, respectively), and primary site was less frequently parameningeal in infants compared with older children (3% v 14%, respectively; P = .001). Local relapse rates were not different between infants, children, and adolescents treated in the same protocols (relapse rates: less than 1 year, 32%; 95% CI, 20% to 46%; 1 to 9 years, 32%; 95% CI, 28% to 36%; and 10 years, 36%; 95% CI, 28% to 46%). No difference in outcome was observed between the two groups of localized RMS tumors (Fig 5). Five-year OS and EFS rates were 72% and 57% in infants versus 69% and 55% in older children, respectively (P = not significant for both). When compared with older children, infants and children had the same prognosis and a better outcome than adolescents (Table 6). Survival for infants with localized alveolar RMS (5-year OS and EFS, 37% and 21%, respectively) was as poor as for older children (5-year OS and EFS, 41% and 30%, respectively). For non-RMS MMT, the overall prognosis was statistically comparable between infants and older children (75% in infants and 71% in older children, P = .61).
DISCUSSION
This large SIOP study reviewed the data of 102 infants diagnosed with MMT during their first year of life and treated according to the MMT 84 and MMT 89 protocols. The largest individual diagnostic subgroup was RMS and, as reported for older children with localized RMS, a male predominance (sex ratio: 1.9) was observed. Despite a high rate of alveolar histology and primary limb tumors, both of which are associated with a poor prognosis,15 the 5-year OS rate for patients with nonmetastatic RMS (72%) was as good as that for older children treated during the same period. Despite the large number of infants with fibrosarcoma, the results were also identical when non-RMS MMT were compared according to age. The absence of an adverse impact of age has been reported in other series including both RMS and non-RMS MMT in infants,3,7 and one study of non-RMS MMT concluded that infants had a better prognosis than older children.6 This small study comprising only 13 infants included six patients with congenital fibrosarcoma, which is known to have a favorable prognosis. Nevertheless, recent reports have found that age less than 1 year is a poor independent prognostic factor in RMS.8,9,13 However, some of these series9 included undifferentiated sarcomas, which are known to have a poorer prognosis,7 in the group of infants with RMS and analyzed failure-free survival. Infants are known to have a higher relapse rate than older children probably because they receive less local therapy (surgery and radiotherapy) or chemotherapy than the usual treatment administered to older children.8,9 Since the first protocols, the SIOP has adopted a conservative approach in terms of local therapy in children with RMS,15,16 which may result in more local relapse than other groups who apply systematic local therapy.7 In this series, the relapse rate was the same for all groups of children (< 1 year, 1 to 9 years, and 10 years old), but the salvage gap (46% of children alive in remission after relapse with second-line treatment) is quite satisfactory in infants, and the OS is comparable to that of older children. Most relapses are local, and children who were successfully treated with salvage second-line treatment had not previously received radiotherapy or large local surgery, which may explain the good salvage rate in this population. This conservative policy, which is designed to avoid or minimize the burden of late effects, is also acceptable in infants. In our experience, 78% of survivors with localized RMS were cured without large or radical therapy.
As previously observed in older children,15 patients with alveolar RMS in the present study had a much poorer outcome than those with embryonal tumors or non-RMS MMT. Interestingly, Grundy et al22 suggested that the behavior and molecular biology of congenital alveolar RMS may be different from those observed in older children with the same histologic subtype. They reported the absence of characteristic translocations in all four of their children and indicated that none of the 10 patients with congenital alveolar RMS reported in the literature at that time (including their four patients) had survived. In our series, only one infant less than 1 month old had alveolar RMS, and this infant died despite treatment. Alveolar histology was also seen more frequently in infants with orbital and nonparameningeal tumors than in older children with these same sites. OS for infants with embryonal RMS and non-RMS MMT (5-year OS rates, 79% and 70%, respectively) was comparable, although the favorable outcome expected for the relatively large number of fibrosarcomas and undifferentiated spindle-cell sarcomas in the non-RMS MMT group would be expected to enhance overall prognosis.23 In our series, as previously described, infants with infantile fibrosarcoma or undifferentiated spindle-cell tumors achieved good survival6,11 in response to conservative treatment in most cases. A relatively large group of patients had a diagnosis of undifferentiated sarcoma, but immunohistochemical staining was less extensive, particularly in the MMT 84 protocol, and no systematic cytogenetic examination of the tumors was performed. In the future, improved diagnostic techniques are expected to refine the diagnostic classification.
Only approximately 2% of childhood sarcomas are diagnosed during the neonatal period.11 The outcome of the newborn group in this study (5-year OS rate, 67%) was satisfactory despite the limitations concerning both chemotherapy and local treatment, but this study included only a small number of patients. Most other data at this age are derived from experiences with small numbers of patients.2,11,22,24 A multi-institutional study conducted by the Children’s Cancer Group, including 32 infants with neonatal MMT (diagnosed during the first month of life), reported a 100% survival rate in 12 infants with fibrosarcoma but nine deaths among 11 infants with RMS.11 An Intergroup Rhabdomyosarcoma Study reported a 3-year OS rate of 49% in 14 infants with neonatal RMS.12 In summary, one half to two thirds of infants with MMT survived.2,24
There is evidence to suggest that infants less than 1 year of age receiving full doses of some drugs, such as vincristine, dactinomycin, or doxorubicin, have an increased incidence of severe toxicity and toxic deaths compared with older children.14 The reasons for their increased sensitivity to cancer chemotherapy are unclear but are probably related to the immaturity of organ systems at this age. In our experience, chemotherapy toxicity was able to be managed by appropriate dose reduction. Nevertheless, this toxicity may have been slightly underreported in our series because of the fact that toxicity data were only recorded for infants with RMS included in the protocol. In our experience, no veno-occlusive disease was observed after chemotherapy in infants. For unknown reasons, the IVA combination induces a low rate of veno-occlusive disease in children (infants or children, MMT 95 protocol, personal data) compared with other dactinomycin-containing regimens. The single patient who died from chemotherapy toxicity had received full-dose dactinomycin in violation of the protocol. Acute tubular and glomerular toxicity after ifosfamide administration is not uncommon in infants (10% to 20%), and precautions must be taken to avoid this drug or to reduce dosages in young infants.
This series confirms the proposal for the future European localized RMS protocol for children less than 1 year old, with drug doses calculated by weight (ie, usual dose in m2 divided by 30). Infants less than 1 month old will receive only VA. Between 1 and 3 months of age, a further reduction of ifosfamide (ie, 50% for the first course) will be adopted, and the dose will then be gradually increased during subsequent courses when the first courses are well tolerated. Anthracyclines will be avoided in infants less than 3 months of age.8 In infants less than 6 months of age, drug doses will be calculated by weight and then progressively increased for subsequent courses to the usual dose in meters squared.
In summary, this study reports that outcome for infants with MMT can be as satisfactory as that achieved in older children, despite the necessary limitations concerning intensity of therapy. The prognosis for infants with alveolar RMS and metastatic disease is as poor as for older children, and these infants should receive intensified chemotherapy and aggressive local therapy.
Appendix
Patients were entered onto the study by treatment centers in the following cities: Argentina: Buenos Aires; Belgium: Bruxelles (Fabiola), Bruxelles (St Joseph), Bruxelles (St Luc), Leuven, Montegnee; Denmark: Aarhus, Copenhagen; France: Angers, Brest, Besancon, Bordeaux, Bordeaux (Bergonie), Caen, Caen (Baclesse), Clermont Ferrand, Colmar, Grenoble, Lille (Centre Anti Cancereux), Lille (Calmette), Lille (St Anthony), Limoges, Lyon, Marseille, Montpelier, Nancy, Nantes, Nice, Paris (Curie), Paris (Trousseau), Poitiers, Reims, Rennes, Rouen, Rouen (Becquerel), St Etienne, Strasbourg (Centre Hospitalier Universitaire [CHU]), Strasbourg (Hautepierre), Toulouse (CHU), Tours, Villejuif; Republic of Ireland: Dublin; the Netherlands: Amsterdam (Academic Medical Center), Amsterdam (Vrije Universiteit), Nijmege; Poland: Wroclaw; Spain: Badalona, Barcelona (Val d’Hebron), Santiago de Compostela, Las Palmas, Madrid (La Paz), Madrid (Nio Jesus), Madrid (12 Octubre), Madrid (La Zarzuela), Oviedo, Pamplona (Virgen del Camino), Valencia (La Fe), Vizcaya (Cruces), Zaragoza (Miguel Servet); Switzerland: Basel, Lausanne, Lucerne, Zurich; and United Kingdom: Belfast, Birmingham, Bristol, Cambridge, Cardiff, Edinburgh, Glasgow, Leeds, Leicester, Liverpool, London (Great Ormond Street), London (Middlesex), London (St Bartholomew’s), Manchester, Newcastle, Nottingham, Oxford, Sheffield, Southampton, Sutton (Royal Marsden).
Authors’ Disclosures of Potential Conflicts of Interest
The authors indicated no potential conflicts of interest.
NOTES
Authors’ disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
Coffin CM, Dehner LP: Soft tissue tumors in first year of life: A report of 190 cases. Pediatr Pathol 10:509-526, 1990
Campbell AN, Chan HS, O’Brien A, et al: Malignant tumours in the neonate. Arch Dis Child 62:19-23, 1987
Ragab AH, Heyn R, Tefft M, et al: Infants younger than 1 year of age with rhabdomyosarcoma. Cancer 58:2606-2610, 1986
Isaacs H Jr: Perinatal (congenital and neonatal) neoplasms: A report of 110 cases. Pediatr Pathol 3:165-216, 1985
Isaacs H Jr: Congenital and neonatal malignant tumors: A 28-year experience at Children’s Hospital of Los Angeles. Am J Pediatr Hematol Oncol 9:121-129, 1987
Hayes-Jordan AA, Spunt SL, Poquette CA, et al: Nonrhabdomyosarcoma soft tissue sarcomas in children: Is age at diagnosis an important variable J Pediatr Surg 35:948-953, 2000
Crist WM, Anderson JR, Meza JL, et al: Intergroup rhabdomyosarcoma study: IV. Results for patients with nonmetastatic disease. J Clin Oncol 19:3091-3102, 2001
Ferrari A, Casanova M, Bisogno G, et al: Rhabdomyosarcoma in infants younger than one year old: A report from the Italian Cooperative Group. Cancer 97:2597-2604, 2003
Joshi D, Anderson JR, Paidas C, et al: Age is an independent prognostic factor in rhabdomyosarcoma: A report from the Soft Tissue Sarcoma Committee of the Children’s Oncology Group. Pediatr Blood Cancer 42:64-73, 2004
Koscielniak E, Harms D, Schmidt D, et al: Soft tissue sarcomas in infants younger than 1 year of age: A report of the German Soft Tissue Sarcoma Study Group (CWS-81). Med Pediatr Oncol 17:105-110, 1989
Dillon PW, Whalen TV, Azizkhan RG, et al: Neonatal soft tissue sarcomas: The influence of pathology on treatment and survival—Children’s Cancer Group Surgical Committee. J Pediatr Surg 30:1038-1041, 1995
Lobe TE, Wiener ES, Hays DM, et al: Neonatal rhabdomyosarcoma: The IRS experience. J Pediatr Surg 29:1167-1170, 1994
Stiller CA, Stevens MC, Magnani C, et al: Survival of children with soft-tissue sarcoma in Europe since 1978: Results from the EUROCARE study. Eur J Cancer 37:767-774, 2001
Morgan E, Baum E, Breslow N, et al: Chemotherapy-related toxicity in infants treated according to the Second National Wilms’ Tumor Study. J Clin Oncol 6:51-55, 1988
Flamant F, Rodary C, Rey A, et al: Treatment of non-metastatic rhabdomyosarcomas in childhood and adolescence: Results of the second study of the International Society of Paediatric Oncology—-MMT 84. Eur J Cancer 34:1050-1062, 1998
Sanchez de Toledo JOO, Bouvet-Forteau N, Praquin M-T, et al: Treatment of non metastatic rhabdomyosarcoma (RMS): Final report from the SIOP MMT 89 study. Med Pediatr Oncol 37:193, 2001 (abstr)
Carli M, Colombatti R, Oberlin O, et al: High-dose melphalan with autologous stem-cell rescue in metastatic rhabdomyosarcoma. J Clin Oncol 17:2796-2803, 1999
Donaldson SS, Draper GJ, Flamant F, et al: Topography of childhood tumors: Pediatric coding system. Pediatr Hematol Oncol 3:249-258, 1986
Rodary C, Flamant F, Donaldson SS: An attempt to use a common staging system in rhabdomyosarcoma: A report of an international workshop initiated by the International Society of Pediatric Oncology (SIOP). Med Pediatr Oncol 17:210-215, 1989
Newton WA Jr, Gehan EA, Webber BL, et al: Classification of rhabdomyosarcomas and related sarcomas: Pathologic aspects and proposal for a new classification—An Intergroup Rhabdomyosarcoma Study. Cancer 76:1073-1085, 1995
Skinner R, Sharkey IM, Pearson AD, et al: Ifosfamide, mesna, and nephrotoxicity in children. J Clin Oncol 11:173-190, 1993
Grundy R, Anderson J, Gaze M, et al: Congenital alveolar rhabdomyosarcoma: Clinical and molecular distinction from alveolar rhabdomyosarcoma in older children. Cancer 91:606-612, 2001
Kynaston JA, Malcolm AJ, Craft AW, et al: Chemotherapy in the management of infantile fibrosarcoma. Med Pediatr Oncol 21:488-493, 1993
Parkes SE, Muir KR, Southern L, et al: Neonatal tumours: A thirty-year population-based study. Med Pediatr Oncol 22:309-317, 1994(D. Orbach, A. Rey, O. Obe)
Institut Gustave Roussy, Villejuif, France
Hospital Val d’Hebron, Barcelona, Spain
Jeroen Bosch Ziekenhuis, ’s-Hertogenbosch, the Netherlands
Royal Hospital for Children, Bristol, United Kingdom
ABSTRACT
PURPOSE: To describe the outcome of infants with a histologically confirmed diagnosis of malignant mesenchymal tumor (MMT) included in the International Society of Paediatric Oncology studies MMT 84 and MMT 89.
PATIENTS AND METHODS: One hundred two infants ( 12 months old) were included. Twenty-four children were less than 3 months old, and 16 were less than 1 month old. Sixty-four patients had rhabdomyosarcoma (RMS), 26 had undifferentiated sarcoma, and 12 had other histology. Clinical TNM stage was stage I (41%), II (39%), III (6%), and IV (14%). First-line treatment was ifosfamide, vincristine, dactinomycin, whereas the second-line combination consisted of either cisplatin and doxorubicin (in MMT 84) or vincristine, carboplatin, etoposide/teniposide (in MMT 89). Chemotherapy doses were adapted to age. Local therapy was conservative surgery as often as possible.
RESULTS: After a median follow-up of 7.8 years (range, 0.1 to 13 years), 5-year overall survival (OS) and event-free survival rates were 66% and 55% for the total study population and 72% and 60% for nonmetastatic patients, respectively. Only two of 13 stage IV patients survived. Sixty-seven percent of newborn infants survived. Infants with alveolar subtype had a poorer survival than those with non-RMS MMT or nonalveolar RMS (5-year OS, 37% v 75% or 82%, respectively; P = .002). When compared with older children with MMT, young age does not seem to be an important prognostic factor.
CONCLUSION: OS was satisfactory even when local treatment was not aggressive, although the prognosis was poor for infants with alveolar RMS or metastatic tumors. Chemotherapy toxicity was manageable with appropriate dose modification.
INTRODUCTION
More than 75% of soft tissue tumors diagnosed between birth and 12 months of age are pathologically benign1 and include diagnoses such as infantile hemangioendothelioma, lymphangioma, or infantile myofibromatosis. Ten percent of tumors are borderline lesions, and only 15% of such tumors in this age group are malignant. Previous reports suggest that the main specific diagnoses of malignant mesenchymal tumors (MMTs) are rhabdomyosarcoma (RMS) and peripheral primitive neuroectodermal tumor,1 although a significant proportion of these infantile tumors are undifferentiated.2-5 Overall, 5% to 11% of childhood MMTs are reported in infants before the age of 1 year.6-9 Prognosis depends on histology, although, for the same diagnosis, the prognosis may be more favorable in infants than in older children. For example, survival rates as high as 100% have been reported in congenital fibrosarcoma.10,11 However, the outlook for neonatal RMS has been reported to be less favorable than for older children.12 In general, age seems to be a prognostic factor in RMS.3,6-8 Recent studies have shown that infants with RMS have a less satisfactory outcome than older children.8,9,13 Clinical management remains difficult in infants because of the immaturity of many physiologic systems.14 The risk of long-term sequelae, particularly after radical surgery, radiotherapy, or the use of alkylating agent chemotherapy, remains a constant preoccupation in the choice of treatment2 at this age. We report the characteristics and outcome of a group of 102 infants less than 13 months old with MMTs registered onto the International Society of Pediatric Oncology (SIOP) MMT 8415 and MMT 8916 protocols, and we compare these results with the results observed in older children treated in the same studies.
PATIENTS AND METHODS
Inclusion Criteria
Infants were prospectively registered onto the MMT 84 and MMT 89 studies from 1984 to 1995. Diagnoses were confirmed by the SIOP Pathology Review Committee. The MMT 84 protocol registered children diagnosed between the ages of 3 months and 18 years, whereas the MMT 89 protocol included children from birth to 18 years. Children aged more than 3 months who were registered in MMT 89 and who had metastatic disease were treated according to a separate European Intergroup study.17 According to European law at the time of the studies, no informed consent was required for enrollment onto these studies. Data on a total of 102 infants (20 registered onto MMT 84 and 82 registered onto MMT 84) were analyzed for this report.
Exclusion Criteria
All children with a reviewed diagnosis indicating a benign tumor (fibromatosis, inflammatory tumor, hamartoma, or myxoma/myxofibroma) or children with tumors that could not be confirmed histologically were excluded from this study. Primary tumor site was classified into the following six standard subgroups: nonparameningeal orbit; parameningeal head and neck (including orbit with parameningeal extension); nonparameningeal head and neck; genitourinary, including bladder/prostate or nonbladder/prostate (paratesticular or vaginal/uterine sites); limbs; and other sites.18
Clinical staging was defined according to the SIOP–International Union Against Cancer TNM system.19 Postsurgical staging was defined in the protocol by the SIOP pathologic TNM staging system, but these data were also presented according to the Intergroup Rhabdomyosarcoma Study group.
Histologic Classification
Since 1984, all tumors have been reviewed by an international panel of pathologists and classified as RMS (according to the new RMS international classification20) or as non-RMS MMT. The diagnostic classification did not take into account any data from cytogenetic or molecular genetic studies on individual tumors. Undifferentiated sarcomas composed of spindle cells were considered separately from the other undifferentiated tumors and were assessed together with fibrosarcomas because of their similar behavior.
Treatment
The details of the MMT 84 and MMT 89 protocols have been published previously.15,16 The strategy used for stage I and II disease was determined by the extent of primary surgery. Patients with complete primary resection (group 1) received chemotherapy consisting of three courses of ifosfamide, vincristine, and dactinomycin (IVA) in MMT 84 but received only four courses of vincristine and dactinomycin (VA) in MMT 89 (Table 1). Incomplete primary resection (groups 2 and 3) was followed by six to 10 courses of IVA (ifosfamide 6 g/m2/course) in MMT 84 and by six courses of IVA in MMT 89 (ifosfamide 9 g/m2/course). In cases of partial response or progressive disease, second-line chemotherapy consisted of cisplatin and doxorubicin in MMT 84 and vincristine, carboplatin, and teniposide (or etoposide) in MMT 89. To minimize late effects, further local therapy was administered only to patients who did not achieve complete remission (CR) with primary surgery and/or subsequent chemotherapy. Whenever possible, this was limited to conservative surgery, and radiotherapy was avoided in these young patients. No specific guidelines concerning radiotherapy (dose or field) were specified apart from the use of brachytherapy whenever possible. Significant local therapy was defined as either large (major surgery but with no predictable major long-term functional or cosmetic consequences and/or external-beam radiotherapy) or radical (major surgery with predictable functional or cosmetic consequences; eg, total cystectomy, orbital exenteration, and amputation).
All patients with either metastatic, node-positive or parameningeal tumors registered onto the MMT 89 protocol were treated with an intensified six-drug chemotherapy regimen, which included IVA; carboplatin, epirubicin, and vincristine; and ifosfamide, vincristine, and etoposide (Table 1).
In both studies, chemotherapy doses were reduced by 50% in infants under the age of 6 months and by 33% in infants between the ages of 6 and 12 months. When the first course was well tolerated, chemotherapy was increased to full dose for the following courses. Infants less than 1 month old initially received alternative single-agent therapy with VA at 50% of the standard dose.
Toxicity
Chemotherapy toxicity was classified into five grades (grade 0 = no toxicity; grade 4 = life-threatening toxicity) according to the National Cancer Institute Common Toxicity Criteria (version 1) and a specific grading system for nephrotoxicity.21 Toxicity data were systematically and prospectively assessed in children included in the MMT 89 protocol. Long-term complications (local or systemic) were not systematically recorded.
Statistical Methods
Outcome was defined by overall survival (OS) and event-free survival (EFS). For EFS, events were defined as relapse after CR or death from any cause (including progression of disease without achieving CR). If a patient failed to achieve CR, the time-point for failure was the date of death (date of progression was not recorded). This method of calculation, in children who died without achieving remission, does not alter the 5-year EFS rate but may slightly alter the slope of the survival curve over the first few years.
Local control was defined as control of the primary tumor site with disappearance of all clinical and radiologic signs of disease or stable residual radiographic images for 6 months after completion of treatment. Statistical analysis was performed using a general database management system. Survival curves were calculated by the Kaplan-Meier method. The statistical significance of each variable was first tested by the log-rank test (univariate procedure). Some data were then compared with the data of older RMS or non-RMS children (aged 1 to 18 years) included and treated in the SIOP MMT 84 and 89 protocols during the same period.
RESULTS
Patient Population
One hundred two infants under the age of 13 months, who were treated between 1984 and 1995, were analyzed in this study. Most of these infants were registered after 1990 (20 patients in the MMT 84 protocol and 82 in the MMT 89 protocol). This group represented 8% of all children (102 of 1,228 children) entered onto the two studies. Median age at diagnosis was 5 months (range, 0 to 12 months); 16 infants were less than 1 month old, and 24 were less than 3 months old. The sex ratio was 1.9 (67 males and 35 females). Median follow-up time of the whole study population was 7.8 years (range, 0.1 to 13 years).
Histologic Classification
Diagnoses as determined by the pathology review committee are listed in Table 2. RMS tumors were more frequent than non-RMS tumors (64 v 38 tumors). The alveolar subtype was detected in 31% of infants with RMS. Undifferentiated sarcoma was the most frequent non-RMS histologic type.
Primary tumor sites according to histology are listed in Table 3. The primary tumor involved the limbs in 30 patients, corresponding to 20% of embryonal RMS, 30% of alveolar RMS, and 39% of non-RMS. Other sites, such as pelvic, abdominal, and retroperitoneal sites, were frequent and mostly concerned non-RMS MMT. Most of the genitourinary tumors were embryonal RMS. Only four parameningeal tumors and four orbital tumors were observed, but three of the four orbital tumors showed alveolar histology.
Clinical Staging
As shown in Table 4, most tumors were initially localized (stage I or II), although 14% of patients presented with distant metastatic disease at diagnosis. Primary surgical resection was complete in only 13% of patients and resulted in microscopic residual disease in 23%. The majority of patients had macroscopic residual disease after the initial surgical resection or biopsy. An unexpectedly high proportion of children (12%) presented with associated pleural or peritoneal effusion (stage pT3c).
CR was obtained in 77% of the infants (76 of 98 infants) for whom this information was available. The CR rate varied according to stage of disease; the CR rate was 95% in stage I patients (39 of 41 patients), 74% in stage II patients (29 of 39 patients), 57% in stage III patients (four of seven patients), and 36% in stage IV patients (four of 11 patients). Data were missing for four patients. Only 13% of patients achieved CR with initial surgery alone (58% after incomplete initial surgery plus chemotherapy, 21% after chemotherapy followed by secondary surgery, and 7% after chemotherapy ± surgery plus radiotherapy; 3% received brachytherapy). One infant with stage IV disease achieved CR after chemotherapy, radiotherapy, surgery, and high-dose chemotherapy. The treatment received to achieve CR was not reported in one child. Although radiotherapy was not recommended regardless of age, five infants received this modality as first-line treatment based on the clinician’s preference. These infants had a median age of 8 months (range, 5 to 12 months) and had stage II disease (embryonal RMS, n = 4; and embryonal sarcoma, n = 1). Treatment was delivered either by external-beam radiotherapy (44 to 45 Gy; n = 3) to the initial tumor volume with margins (n = 2) or to the residual tumor volume after chemotherapy (n = 1) or by brachytherapy (50 to 65 Gy; n = 2). None of these infants developed local relapse.
Chemotherapy Toxicity
Detailed toxicity data were available only for patients on the MMT 89 protocol. Table 5 lists the percentages of children experiencing grade 3 or 4 toxicity after chemotherapy for whom complete data were available. Toxicity was mainly hematologic and less commonly infectious. This toxicity led to an additional dose reduction in 19% of infants receiving IVA; 33% of infants receiving vincristine, carboplatin, and teniposide (or etoposide); 43% of infants receiving carboplatin, epirubicin, and vincristine; and 60% of infants receiving ifosfamide, vincristine, and etoposide; no reductions were necessary after VA. Five infants who received ifosfamide experienced severe acute renal toxicity (glomerular, n = 1; or tubular, n = 4). These infants were 1, 2, 7, 9, and 12 months old at the time of this toxicity.
Relapse
Twenty-six (34%) of the 76 infants who achieved CR subsequently relapsed a median of 13 months (range, 3 to 40 months) after the start of treatment. Relapse was mainly local (69%), but relapse was metastatic in 23% of patients and local and metastatic in 8% of patients. Further treatment consisted of chemotherapy alone (n = 3), surgery alone (n = 1), chemotherapy and surgery (n = 4), chemotherapy with radiotherapy ± surgery (n = 9), or surgery and radiotherapy (n = 2). No treatment details were available for seven patients. At final analysis, 12 (46%) of 26 relapsing patients with embryonal RMS (n = 4), alveolar RMS (n = 2), undifferentiated spindle-cell sarcoma (n = 1), other undifferentiated sarcomas (n = 3), or non-RMS sarcomas (n = 2) were alive with a median follow-up of 7 years (range, 1 to 10 years) after relapse.
Cause of Death
Thirty-five infants died, mainly from their disease (n = 33). One infant died from toxicity (a 5-month-old infant in whom the dosage of dactinomycin was not correctly reduced according to age and who developed liver failure). One child, who was treated 33 months previously for alveolar RMS, died after treatment for a second malignancy (acute myeloid leukemia).
Survival
The outcome of the whole group showed a 5-year OS rate of 66% and a 5-year EFS rate of 55% (Fig 1). More favorable results were observed in localized tumors (stages I to III: OS, 73% and EFS, 60%; Fig 2) than in metastatic disease (OS and EFS, 17%; 95% CI, 5% to 45%). Survival was better for infants with localized nonalveolar RMS or non-RMS MMT than for infants with localized alveolar RMS (5-year OS, 82%, 75%, and 40%, respectively; P = .004; Fig 3). Similar results were observed for EFS (Fig 4).
Newborns
Sixteen infants were less than 1 month old at diagnosis (10 males and six females). The histologic distribution was as follows: five RMS (one alveolar), two undifferentiated spindle-cell sarcomas, one fibrosarcoma, one peripheral primitive neuroectodermal tumor, one embryonal sarcoma, and six other non-RMS MMT. The primary site was the limbs in eight patients; others sites included nonparameningeal head and neck (n = 3), genitourinary (n = 1), and other sites (n = 4). Clinical stages were as follows: stage I (n = 5), II (n = 9), and IV (n = 2). CR was obtained in nine (64%) of the 14 assessable patients and was achieved with chemotherapy alone in six patients, chemotherapy and surgery in one patient, and surgery alone in two patients (missing data for two patients). Eleven of the 16 infants are alive, with a follow-up ranging from 1.8 to 10 years, and the 5-year OS rate was 67%. Three of the five infants with RMS are alive.
Patients With Metastatic Disease
Thirteen infants (six males and seven females) had metastatic disease at the time of diagnosis (alveolar RMS, n = 5; nonalveolar RMS, n = 3; and undifferentiated sarcomas, n = 5). The primary sites were the limbs (n = 3), parameningeal head and neck (n = 2), paratesticular/vaginal (n = 2), orbit (n = 1), and other sites (n = 5). Metastases were usually multiple and involved bone (n = 3), lungs (n = 3), bone marrow (n = 2), liver (n = 2), CNS (n = 1), or other sites (n = 6). Only three infants achieved CR (with chemotherapy; chemotherapy plus surgery; and chemotherapy, surgery, radiotherapy, and high-dose chemotherapy). Ten of the 12 assessable infants died (one was lost to follow-up) after a median follow-up of 11 months. Only two infants, one with embryonal RMS and one with undifferentiated sarcoma and initial bone and liver metastases, are still alive 81 and 85 months after diagnosis, respectively.
Comparison Between Histologic Types
Fibrosarcoma and undifferentiated spindle-cell sarcoma represented the largest specified diagnostic group among non-RMS MMT. Twelve infants (eight males and four females) had fibrosarcoma (n = 5) or undifferentiated sarcomas composed of spindle cells (n = 7). The age at diagnosis ranged from 7 days to 4 months. The primary site was usually the limbs (n = 7), with other sites in five infants. The initial staging was as follows: stage I (n = 6), II (n = 4), III (n = 1), and unknown (n = 1). When known, CR was obtained in eight of nine patients with chemotherapy (five patients: six courses of IVA; one patient: three courses of IVA; and one patient: three courses of IVA and three courses of cisplatin and doxorubicin) and/or conservative surgery (after chemotherapy in two patients or alone in one patient). In two patients, the residual mass progressively disappeared with time after chemotherapy or surgery alone (missing data for one patient). OS and 5-year EFS rates were 92% (95% CI, 65% to 99%) and 83% (95% CI, 55% to 95%), respectively.
Undifferentiated sarcoma with no pattern of differentiation (other undifferentiated sarcoma) was observed in 19 infants (13 males and six females). Six infants were less than 1 month old at diagnosis, two were aged between 1 and 3 months, two were aged between 3 and 6 months, and nine were older than 6 months. The primary site was mostly other sites (n = 8) or limbs (n = 5). Initial staging was stage I (n = 5), II (n = 8), and IV (n = 5), with missing data for one patient. CR was obtained in nine of 12 patients, when data were available, with surgery only (n = 1), chemotherapy (n = 5), or chemotherapy followed by surgery (n = 3). At 5 years, the OS and EFS rates were 56% (95% CI, 35% to 46%) and 45% (95% CI, 25% to 67%).
The median age of the 64 infants (42 males and 22 females) with RMS was 7 months. The primary sites were as follows: limbs (n = 15), genitourinary nonbladder/prostate (n = 13), genitourinary bladder/prostate (n = 12), head and neck nonparameningeal (n = 8), head and neck parameningeal (n =2), orbit (n = 4), and other sites (n = 10). CR was achieved in 87% of patients. Initial first CR for infants with localized RMS was obtained with surgery only (14%), chemotherapy (50%), chemotherapy followed by conservative surgery (27%), chemotherapy and brachytherapy (3%), chemotherapy followed by surgery and brachytherapy (3%), or chemotherapy, surgery, and external radiotherapy (3%). The EFS and OS rates according to histologic subtype in localized tumors are indicated in Figures 3 and 4. Among the 31 infants alive with a minimum follow-up of 2 years who were treated in the MMT 89 protocol for localized RMS, four were treated by chemotherapy alone, 20 were treated by chemotherapy and conservative surgery, and seven were treated by chemotherapy with significant local therapy. Only two patients were treated with local therapy defined as large, and five were treated with local therapy defined as radical. In summary, 78% of survivors were treated using a local conservative approach.
Comparison With Older Patients
The characteristics and outcome of infants with RMS were compared with those of a group of older children with RMS (age range, 1 to 18 years) who were treated in the same studies. Sex ratio, clinical stage, and CR rate were not significantly different. There was a higher proportion of alveolar RMS subtypes in the infant group than in the older children (31% v 19%, respectively; P = .01). Microscopically incomplete initial surgery was more frequent in infants than in older children (31% v 15%, respectively; P < .01). Primary site was more often genitourinary bladder/prostate and limbs in infants (19% v 9% in older children; and 23% v 15% in older children, respectively), and primary site was less frequently parameningeal in infants compared with older children (3% v 14%, respectively; P = .001). Local relapse rates were not different between infants, children, and adolescents treated in the same protocols (relapse rates: less than 1 year, 32%; 95% CI, 20% to 46%; 1 to 9 years, 32%; 95% CI, 28% to 36%; and 10 years, 36%; 95% CI, 28% to 46%). No difference in outcome was observed between the two groups of localized RMS tumors (Fig 5). Five-year OS and EFS rates were 72% and 57% in infants versus 69% and 55% in older children, respectively (P = not significant for both). When compared with older children, infants and children had the same prognosis and a better outcome than adolescents (Table 6). Survival for infants with localized alveolar RMS (5-year OS and EFS, 37% and 21%, respectively) was as poor as for older children (5-year OS and EFS, 41% and 30%, respectively). For non-RMS MMT, the overall prognosis was statistically comparable between infants and older children (75% in infants and 71% in older children, P = .61).
DISCUSSION
This large SIOP study reviewed the data of 102 infants diagnosed with MMT during their first year of life and treated according to the MMT 84 and MMT 89 protocols. The largest individual diagnostic subgroup was RMS and, as reported for older children with localized RMS, a male predominance (sex ratio: 1.9) was observed. Despite a high rate of alveolar histology and primary limb tumors, both of which are associated with a poor prognosis,15 the 5-year OS rate for patients with nonmetastatic RMS (72%) was as good as that for older children treated during the same period. Despite the large number of infants with fibrosarcoma, the results were also identical when non-RMS MMT were compared according to age. The absence of an adverse impact of age has been reported in other series including both RMS and non-RMS MMT in infants,3,7 and one study of non-RMS MMT concluded that infants had a better prognosis than older children.6 This small study comprising only 13 infants included six patients with congenital fibrosarcoma, which is known to have a favorable prognosis. Nevertheless, recent reports have found that age less than 1 year is a poor independent prognostic factor in RMS.8,9,13 However, some of these series9 included undifferentiated sarcomas, which are known to have a poorer prognosis,7 in the group of infants with RMS and analyzed failure-free survival. Infants are known to have a higher relapse rate than older children probably because they receive less local therapy (surgery and radiotherapy) or chemotherapy than the usual treatment administered to older children.8,9 Since the first protocols, the SIOP has adopted a conservative approach in terms of local therapy in children with RMS,15,16 which may result in more local relapse than other groups who apply systematic local therapy.7 In this series, the relapse rate was the same for all groups of children (< 1 year, 1 to 9 years, and 10 years old), but the salvage gap (46% of children alive in remission after relapse with second-line treatment) is quite satisfactory in infants, and the OS is comparable to that of older children. Most relapses are local, and children who were successfully treated with salvage second-line treatment had not previously received radiotherapy or large local surgery, which may explain the good salvage rate in this population. This conservative policy, which is designed to avoid or minimize the burden of late effects, is also acceptable in infants. In our experience, 78% of survivors with localized RMS were cured without large or radical therapy.
As previously observed in older children,15 patients with alveolar RMS in the present study had a much poorer outcome than those with embryonal tumors or non-RMS MMT. Interestingly, Grundy et al22 suggested that the behavior and molecular biology of congenital alveolar RMS may be different from those observed in older children with the same histologic subtype. They reported the absence of characteristic translocations in all four of their children and indicated that none of the 10 patients with congenital alveolar RMS reported in the literature at that time (including their four patients) had survived. In our series, only one infant less than 1 month old had alveolar RMS, and this infant died despite treatment. Alveolar histology was also seen more frequently in infants with orbital and nonparameningeal tumors than in older children with these same sites. OS for infants with embryonal RMS and non-RMS MMT (5-year OS rates, 79% and 70%, respectively) was comparable, although the favorable outcome expected for the relatively large number of fibrosarcomas and undifferentiated spindle-cell sarcomas in the non-RMS MMT group would be expected to enhance overall prognosis.23 In our series, as previously described, infants with infantile fibrosarcoma or undifferentiated spindle-cell tumors achieved good survival6,11 in response to conservative treatment in most cases. A relatively large group of patients had a diagnosis of undifferentiated sarcoma, but immunohistochemical staining was less extensive, particularly in the MMT 84 protocol, and no systematic cytogenetic examination of the tumors was performed. In the future, improved diagnostic techniques are expected to refine the diagnostic classification.
Only approximately 2% of childhood sarcomas are diagnosed during the neonatal period.11 The outcome of the newborn group in this study (5-year OS rate, 67%) was satisfactory despite the limitations concerning both chemotherapy and local treatment, but this study included only a small number of patients. Most other data at this age are derived from experiences with small numbers of patients.2,11,22,24 A multi-institutional study conducted by the Children’s Cancer Group, including 32 infants with neonatal MMT (diagnosed during the first month of life), reported a 100% survival rate in 12 infants with fibrosarcoma but nine deaths among 11 infants with RMS.11 An Intergroup Rhabdomyosarcoma Study reported a 3-year OS rate of 49% in 14 infants with neonatal RMS.12 In summary, one half to two thirds of infants with MMT survived.2,24
There is evidence to suggest that infants less than 1 year of age receiving full doses of some drugs, such as vincristine, dactinomycin, or doxorubicin, have an increased incidence of severe toxicity and toxic deaths compared with older children.14 The reasons for their increased sensitivity to cancer chemotherapy are unclear but are probably related to the immaturity of organ systems at this age. In our experience, chemotherapy toxicity was able to be managed by appropriate dose reduction. Nevertheless, this toxicity may have been slightly underreported in our series because of the fact that toxicity data were only recorded for infants with RMS included in the protocol. In our experience, no veno-occlusive disease was observed after chemotherapy in infants. For unknown reasons, the IVA combination induces a low rate of veno-occlusive disease in children (infants or children, MMT 95 protocol, personal data) compared with other dactinomycin-containing regimens. The single patient who died from chemotherapy toxicity had received full-dose dactinomycin in violation of the protocol. Acute tubular and glomerular toxicity after ifosfamide administration is not uncommon in infants (10% to 20%), and precautions must be taken to avoid this drug or to reduce dosages in young infants.
This series confirms the proposal for the future European localized RMS protocol for children less than 1 year old, with drug doses calculated by weight (ie, usual dose in m2 divided by 30). Infants less than 1 month old will receive only VA. Between 1 and 3 months of age, a further reduction of ifosfamide (ie, 50% for the first course) will be adopted, and the dose will then be gradually increased during subsequent courses when the first courses are well tolerated. Anthracyclines will be avoided in infants less than 3 months of age.8 In infants less than 6 months of age, drug doses will be calculated by weight and then progressively increased for subsequent courses to the usual dose in meters squared.
In summary, this study reports that outcome for infants with MMT can be as satisfactory as that achieved in older children, despite the necessary limitations concerning intensity of therapy. The prognosis for infants with alveolar RMS and metastatic disease is as poor as for older children, and these infants should receive intensified chemotherapy and aggressive local therapy.
Appendix
Patients were entered onto the study by treatment centers in the following cities: Argentina: Buenos Aires; Belgium: Bruxelles (Fabiola), Bruxelles (St Joseph), Bruxelles (St Luc), Leuven, Montegnee; Denmark: Aarhus, Copenhagen; France: Angers, Brest, Besancon, Bordeaux, Bordeaux (Bergonie), Caen, Caen (Baclesse), Clermont Ferrand, Colmar, Grenoble, Lille (Centre Anti Cancereux), Lille (Calmette), Lille (St Anthony), Limoges, Lyon, Marseille, Montpelier, Nancy, Nantes, Nice, Paris (Curie), Paris (Trousseau), Poitiers, Reims, Rennes, Rouen, Rouen (Becquerel), St Etienne, Strasbourg (Centre Hospitalier Universitaire [CHU]), Strasbourg (Hautepierre), Toulouse (CHU), Tours, Villejuif; Republic of Ireland: Dublin; the Netherlands: Amsterdam (Academic Medical Center), Amsterdam (Vrije Universiteit), Nijmege; Poland: Wroclaw; Spain: Badalona, Barcelona (Val d’Hebron), Santiago de Compostela, Las Palmas, Madrid (La Paz), Madrid (Nio Jesus), Madrid (12 Octubre), Madrid (La Zarzuela), Oviedo, Pamplona (Virgen del Camino), Valencia (La Fe), Vizcaya (Cruces), Zaragoza (Miguel Servet); Switzerland: Basel, Lausanne, Lucerne, Zurich; and United Kingdom: Belfast, Birmingham, Bristol, Cambridge, Cardiff, Edinburgh, Glasgow, Leeds, Leicester, Liverpool, London (Great Ormond Street), London (Middlesex), London (St Bartholomew’s), Manchester, Newcastle, Nottingham, Oxford, Sheffield, Southampton, Sutton (Royal Marsden).
Authors’ Disclosures of Potential Conflicts of Interest
The authors indicated no potential conflicts of interest.
NOTES
Authors’ disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
Coffin CM, Dehner LP: Soft tissue tumors in first year of life: A report of 190 cases. Pediatr Pathol 10:509-526, 1990
Campbell AN, Chan HS, O’Brien A, et al: Malignant tumours in the neonate. Arch Dis Child 62:19-23, 1987
Ragab AH, Heyn R, Tefft M, et al: Infants younger than 1 year of age with rhabdomyosarcoma. Cancer 58:2606-2610, 1986
Isaacs H Jr: Perinatal (congenital and neonatal) neoplasms: A report of 110 cases. Pediatr Pathol 3:165-216, 1985
Isaacs H Jr: Congenital and neonatal malignant tumors: A 28-year experience at Children’s Hospital of Los Angeles. Am J Pediatr Hematol Oncol 9:121-129, 1987
Hayes-Jordan AA, Spunt SL, Poquette CA, et al: Nonrhabdomyosarcoma soft tissue sarcomas in children: Is age at diagnosis an important variable J Pediatr Surg 35:948-953, 2000
Crist WM, Anderson JR, Meza JL, et al: Intergroup rhabdomyosarcoma study: IV. Results for patients with nonmetastatic disease. J Clin Oncol 19:3091-3102, 2001
Ferrari A, Casanova M, Bisogno G, et al: Rhabdomyosarcoma in infants younger than one year old: A report from the Italian Cooperative Group. Cancer 97:2597-2604, 2003
Joshi D, Anderson JR, Paidas C, et al: Age is an independent prognostic factor in rhabdomyosarcoma: A report from the Soft Tissue Sarcoma Committee of the Children’s Oncology Group. Pediatr Blood Cancer 42:64-73, 2004
Koscielniak E, Harms D, Schmidt D, et al: Soft tissue sarcomas in infants younger than 1 year of age: A report of the German Soft Tissue Sarcoma Study Group (CWS-81). Med Pediatr Oncol 17:105-110, 1989
Dillon PW, Whalen TV, Azizkhan RG, et al: Neonatal soft tissue sarcomas: The influence of pathology on treatment and survival—Children’s Cancer Group Surgical Committee. J Pediatr Surg 30:1038-1041, 1995
Lobe TE, Wiener ES, Hays DM, et al: Neonatal rhabdomyosarcoma: The IRS experience. J Pediatr Surg 29:1167-1170, 1994
Stiller CA, Stevens MC, Magnani C, et al: Survival of children with soft-tissue sarcoma in Europe since 1978: Results from the EUROCARE study. Eur J Cancer 37:767-774, 2001
Morgan E, Baum E, Breslow N, et al: Chemotherapy-related toxicity in infants treated according to the Second National Wilms’ Tumor Study. J Clin Oncol 6:51-55, 1988
Flamant F, Rodary C, Rey A, et al: Treatment of non-metastatic rhabdomyosarcomas in childhood and adolescence: Results of the second study of the International Society of Paediatric Oncology—-MMT 84. Eur J Cancer 34:1050-1062, 1998
Sanchez de Toledo JOO, Bouvet-Forteau N, Praquin M-T, et al: Treatment of non metastatic rhabdomyosarcoma (RMS): Final report from the SIOP MMT 89 study. Med Pediatr Oncol 37:193, 2001 (abstr)
Carli M, Colombatti R, Oberlin O, et al: High-dose melphalan with autologous stem-cell rescue in metastatic rhabdomyosarcoma. J Clin Oncol 17:2796-2803, 1999
Donaldson SS, Draper GJ, Flamant F, et al: Topography of childhood tumors: Pediatric coding system. Pediatr Hematol Oncol 3:249-258, 1986
Rodary C, Flamant F, Donaldson SS: An attempt to use a common staging system in rhabdomyosarcoma: A report of an international workshop initiated by the International Society of Pediatric Oncology (SIOP). Med Pediatr Oncol 17:210-215, 1989
Newton WA Jr, Gehan EA, Webber BL, et al: Classification of rhabdomyosarcomas and related sarcomas: Pathologic aspects and proposal for a new classification—An Intergroup Rhabdomyosarcoma Study. Cancer 76:1073-1085, 1995
Skinner R, Sharkey IM, Pearson AD, et al: Ifosfamide, mesna, and nephrotoxicity in children. J Clin Oncol 11:173-190, 1993
Grundy R, Anderson J, Gaze M, et al: Congenital alveolar rhabdomyosarcoma: Clinical and molecular distinction from alveolar rhabdomyosarcoma in older children. Cancer 91:606-612, 2001
Kynaston JA, Malcolm AJ, Craft AW, et al: Chemotherapy in the management of infantile fibrosarcoma. Med Pediatr Oncol 21:488-493, 1993
Parkes SE, Muir KR, Southern L, et al: Neonatal tumours: A thirty-year population-based study. Med Pediatr Oncol 22:309-317, 1994(D. Orbach, A. Rey, O. Obe)