Endocrine late effects in survivors of infantile acute lymphoblastic leukemia

Abstract.

Infantile acute lymphoblastic leukemia (ALL) is a rare disease. In survivors, endocrine late effects, such as growth disorder and hypothyroidism, have been reported, but gonadal function remains unclear. Infantile ALL frequently requires transplantation and higher doses of alkylating agents, even in the absence of transplantation. Some studies in childhood cancer survivors reported that a cyclophosphamide equivalent dose (CED) of > 20 g/m² was associated with testosterone deficiency in boys and > 8 g/m² with ovarian dysfunction in girls. We retrospectively reviewed the treatment and endocrine function of 6 infantile ALL survivors treated at our hospital using their medical records. The patients’ age at the time of the study was between 12 and 26 yr. One patient had 0 transplant, four of them had 1 transplant, and one had 2 transplants, with CEDs of 3, 9–11, and 24 g/m² respectively. Two patients had short stature, and two patients experienced hypothyroidism. All three girls with a CED of 9–11 g/m² had primary hypogonadism, and the boy with a CED of 24 g/m² had high LH and FSH levels, suggesting testosterone deficiency and spermatogenesis disorders. In conclusion, gonadal function, growth and thyroid function should be carefully monitored in infantile ALL, and CED may be useful for predicting the development of hypogonadism.

Highlights

● Higher doses of alkylating agents are warranted for infantile ALL.

● Hypogonadism may be more frequent than short stature or hypothyroidism in infantile ALL survivors.

● CED may be useful in predicting the development of hypogonadism.

Introduction

Infantile acute lymphoblastic leukemia (ALL) is a rare and dismal disease with a 5-yr disease-free survival rate of less than 50% (1). More than 80% of infants with ALL have KMT2A gene rearrangement, which is associated with poor prognosis. Although all patients with KMT2A gene rearrangement were eligible for transplantation as a high-risk group, their prognosis was poor even after transplantation. Late complications are also an important consideration for survivors. Endocrine late effects such as hypothyroidism and short stature have been reported in infantile ALL patients with KMT2A gene rearrangement (Japanese Infantile Leukemia Study Group: JILSG MLL96/98 trial) (2). It is hard to ascertain the gonadal function of infants with ALL as they do not reach puberty; therefore, the number of reports is small and many aspects remain unclear. In general, a protocol combining total-body irradiation (TBI) and alkylating agents is used as a pre-transplant intervention in pediatric cancer transplantation cases. However, since TBI is difficult to perform in infants with developing brains, high-dose alkylating agents, such as busulfan, are commonly used at the time of transplantation for infantile ALL. Therefore, the total dose of alkylating agents tends to be higher in infants with ALL than in children older than one year with ALL. Gonadal function should be a particular concern among late complications in infants with ALL because alkylating agents are generally associated with a risk of gonadal dysfunction. Recently, several studies have reported the relationship between alkylating agent doses used in chemotherapy and gonadal function in children with ALL. Gonadal function was evaluated using the Cyclophosphamide Equivalent Dose (CED), which is calculated by replacing the total dose of alkylating agents with cyclophosphamide (3, 4). It has been reported that a CED of ≥ 4 g/m² in boys is associated with impaired spermatogenesis and that of ≥20 g/m² in boys is associated with testosterone deficiency, whereas a CED of ≥ 8 g/m² in girls is associated with a high risk of decreased ovarian function (4,5,6).

We investigated the endocrine late effects in survivors of infantile ALL at our hospital and have discussed the points to be considered during follow-up.

Patients and Methods

The study was approved by the Medical Research Ethics Committee of Nara Medical University (No. 2826).

Patients

We retrospectively reviewed the treatment, physical status, and endocrine function of six survivors of infantile ALL (three boys and three girls) treated at our hospital between 1996 and 2011, based on electronic medical records. Patients with T-cell ALL were excluded from this study. The age at diagnosis ranged from 1 mo to 11 mo, and 5 of the 6 patients had KMT2A gene rearrangement (Table 1).

Table 1. Patients’ characteristics and treatment for infantile ALL.

Treatments

Treatment of ALL was based on protocols published by the JILSG and the Japanese Pediatric Leukemia and Lymphoma Study Group (JPLSG). One patient was treated with the JILSG MLL96/98 protocol, four patients were treated with the JPLSG MLL03 protocol, and the remaining patient was treated with the JPLSG MLL10 protocol. Five of the six patients underwent hematopoietic stem cell transplantation (HSCT), four patients underwent allogeneic cord blood transplantation, and one patient (Case 6) underwent allogeneic cord blood transplantation at 1 yr of age and allogeneic bone marrow transplantation when relapse occurred at 3 yr of age. TBI was performed only in Case 6, in which the patient relapsed and underwent two transplants, with a total dose of 12 Gy. All patients received only chemotherapy as pre-transplant therapy, except at the time of relapse in Case 6, which included alkylating agents, such as cyclophosphamide, busulfan, and melphalan.

Assessments

Body height and weight were measured at the time of investigation and expressed as the standard deviation (SD) score for each age and sex. Body mass index (BMI) was calculated as weight (kg)/height² (m²), and the BMI z-score was calculated based on BMI reference data for Japanese children (7). Short stature was defined as a height less than –2SD of the standard height or a height velocity less than –1.5SD of the standard value for more than 2 yr. GH deficiency (GHD) was defined as short stature and low peak GH level (< 6 ng/mL) on two or more GH stimulation tests. A BMI z-score of less than 2SD was defined as thin, and a BMI z-score of more than 2SD was defined as obesity. Thyroid function was assessed by measuring the serum TSH and free T4 (FT4) levels. Normal range of TSH was 0.5–5.0 μU/mL, and that of FT4 was 0.9–1.7 ng/dL in our facility. Primary hypothyroidism was defined as TSH ≥10 µIU/mL or TSH 5–9 µIU/mL with FT4 < 0.9 ng/dL, and subclinical hypothyroidism was defined as TSH 5–9 µIU/mL without low FT4 levels. Pubertal status was assessed using Tanner stage 1–5 and by measuring basal serum LH, FSH, and testosterone levels in boys and estradiol (E2) levels in girls. The testicular volume in boys was measured using an orchidometer, and the onset of puberty was defined as volume > 4 mL. Precocious puberty was defined as the onset of pubertal signs in boys under 9 yr and girls under 7.5 yr or as the onset of pubertal signs in boys under 10 yr and girls under 8.5 yr if their height was less than –1SD. Primary hypogonadism was defined as high FSH and LH levels and low testosterone or E2 levels. Adrenocortical function was assessed by ACTH and cortisol levels and in some cases, by urinary cortisol in 24-h urine collection. Glucose and lipid metabolism were evaluated based on hepatic function, serum total cholesterol, LDL cholesterol, fasting triglycerides (TG), uric acid, fasting blood glucose, and HbA1c.

The alkylating agent doses were converted to CED using the previously published equation: CED (mg/m²) = 1.0 (cumulative cyclophosphamide dose [mg/m²]) + 0.244 (cumulative ifosfamide dose [mg/m²]) + 0.857 (cumulative procarbazine dose [mg/m²]) + 14.286 (cumulative chlorambucil dose [mg/m²]) + 15.0 (cumulative BCNU dose [mg/m²]) + 16.0 (cumulative CCNU dose [mg/m²]) + 40 (cumulative melphalan dose [mg/m²]) + 50 (cumulative thio-TEPA dose [mg/m²]) + 100 (cumulative nitrogen mustard dose [mg/m²]) + 8.823 (cumulative busulfan dose [mg/m²]) (3, 4).

Results

This study sought to investigate the endocrine late effects in infantile ALL survivors in our hospital. Their ages at the time of the investigation ranged from 12 to 26 yr. One patient underwent no HSCT, four patients underwent one HSCT, and one patient underwent two HSCTs. The CED was 3 g/m² for no transplant, 9–11 g/m² for one transplant, and 24 g/m² for two transplants. Of these six patients, five had some endocrinological abnormalities. Treatment of endocrinological abnormalities was as follows: two patients (Cases 5 and 6) were treated with levothyroxine for primary hypothyroidism, one patient (Case 3) was treated with sex hormone replacement therapy for primary hypogonadism, and one patient (Case 3) was treated with metformin and basal insulin for type 2 diabetes mellitus (DM) and statin for hyperlipidemia.

Regarding the presence or absence of graft-versus-host disease (GVHD) after HSCT, four patients had acute GVHD (Grade I) with only skin symptoms, and one patient (Case 6) had chronic GVHD (Mild, Limited) (8) (Table 1). Case 5 had a 5-mo history of long-term corticosteroid treatment for idiopathic pneumonia syndrome after HSCT, and Case 6 had a 7-mo history of long-term corticosteroid treatment for chronic GVHD after the second HSCT.

Growth

Height and weight, SD score, BMI z-score, IGF-1 SD score at the time of investigation, and the target height SD scores are shown in Table 2. The growth charts for all cases are shown in Fig. 1. Two cases (Cases 4 and 6) had a short stature of < –2SD (–2.1SD and –3.3SD, respectively). In Case 6, the GH stimulation test was performed, and peak GH levels were 8.59, 8.15, 9.98 ng/mL in arginine, L-DOPA, and clonidine stimulation test, respectively. Furthermore, the patient in Case 6 was also significantly thin, with a BMI z-score of –7.8SD and body fat percentage of 4.8% on dual-energy X-ray absorptiometry images. His trunk adiposity was low at 4.0%, which is expected to indicate low visceral fat mass, and HSCT-associated partial lipodystrophy (9) was ruled out. We are currently providing nutritional guidance to increase the body weight.

Table 2. Results of endocrine late effects.

Fig. 1.

The growth charts of all patients. The height and weight data of the patients were plotted on Japanese standard growth charts for normal Japanese children in 2000. The upper and lower charts show the height and weight, respectively. In some cases, the administration periods of estrogen, levothyroxine, and LH-RH analogs are shown.

Thyroid function

The TSH and FT4 levels at the time of investigation or before the start of thyroid hormone replacement if levothyroxine treatment was administered are shown in Table 2. Two patients (Cases 5 and 6) who had undergone one or two HSCTs were diagnosed with primary hypothyroidism. Case 5 was treated with low-dose levothyroxine since the age of 1 yr and Case 6 since the age of 5 yr, and are currently maintaining a euthyroid status without increasing the dose of levothyroxine.

Gonadal function

All three girls had primary hypogonadism, and the CED was 9–11 g/m² in all cases (Table 2). Although two patients (Cases 2 and 3) were subjected to estrogen replacement therapy with estrogen tablets or estrogen patches, therapy was discontinued in one of them (Case 2) because of abdominal pain at the age of 14 yr. Subsequently, although the LH and FSH levels were high, the E2 values varied from < 10 to 143 pg/mL in Case 2. Since pubertal status was progressing and menstruation was occurring, although irregularly, a growth spurt was observed, bone mineral density was at the lower limit of normal (–2.0SD), and treatment had been halted and she is under observation. However, Kaufman therapy appears to be necessary and will continue to be monitored. In Case 4, gonadotropins and E2 levels also fluctuated, as in Case 2, and partial secretion was thought to remain. No replacement therapy was administered because of the onset of irregular menstruation and bone mineral density within the normal range. In Case 6, with a CED over 20 g/m², the onset of pubertal signs was observed. However, FSH levels were elevated, suggesting impaired spermatogenesis. LH levels were also slightly elevated, suggesting inadequate testosterone production. One patient (Case 5) was treated with gonadal suppression therapy, as he developed puberty when he was 9 yr old, with less than –1SD of his height. His height at the start of the growth spurt was 125 cm, and the final height estimated based on his bone age, if no treatment was given, was 154.4 cm (–2.8SD) by the GP-II method (10). In Case 1, gonadotropin and testosterone levels were not measured due to the end of follow-up. Testosterone production appeared to have been unimpaired because testicular enlargement and a notable growth spurt were observed. However, the patient’s fertility could not be monitored.

Adrenocortical function

None of the patients presented symptoms suggestive of acute adrenal insufficiency between the time of chemotherapy and the time of the survey, and none of the patients required adrenal hormone replacement after treatment with prednisolone. ACTH and cortisol levels were almost normal (Table 2). Although Case 5 had slightly lower cortisol levels (around 6 µg/dL) in the morning, urinary cortisol in 24-hour urine collection was not as low as 32.2 µg/d without symptoms of adrenal insufficiency. Therefore, no supplementation was administered. Further follow-up and, if necessary, further evaluation, such as hormone stimulation tests, will be conducted.

Glucose and lipid metabolism

Glucose and lipid metabolism were evaluated using BMI z-score, hepatic function, serum total cholesterol, LDL cholesterol, fasting triglycerides (TG), uric acid, fasting blood glucose, and HbA1c (Table 2). In Case 3, hepatic function abnormalities, hyperlipidemia, hyperuricemia, and elevated HbA1c were observed despite a lack of obesity, with a BMI z-score of –0.9SD. In this case, at the age of 14, with an HbA1c level of 7.5% and homeostasis model assessment insulin resistance (HOMA-R) of 10.1 by oral glucose tolerance test (OGTT), insulin resistance was observed, and she was diagnosed with type 2 diabetes and started on metformin at 500 mg/d. However, as the HbA1c value continued to increase, the dose of metformin was increased to 1,500 mg/day. At age 16, HbA1c level reached 8.1%, and basal insulin therapy was initiated. In addition, oral therapy with statins had only recently been initiated for hyperlipidemia. Body fat mass was evaluated using dual-energy X-ray absorptiometry. Fat percentage was 38.2% in the upper extremities, 37.8% in the trunk, and 31.1% in the lower extremities, and lower fat mass was observed in the extremities and buttocks than in the trunk and in the abdomen, and she was diagnosed with HSCT associated partial lipodystrophy.

Discussion

In the infantile ALL survivors reviewed in this study, endocrine late complications included growth disorders, hypothyroidism, hypogonadism, and abnormalities in glucose and lipid metabolism. In general, radiation therapy is a risk factor for growth disorder, GH deficiency, hypogonadism, hypothyroidism, and abnormalities in glucose and lipid metabolism in childhood cancer survivors, high-dose alkylating agents are risk factors for hypogonadism, and corticosteroids are risk factors for abnormalities in glucose, lipid, and bone metabolism (11,12,13). According to previous reports on childhood cancer survivors (CCS), including those with solid tumors, endocrine late effects were observed in more than half of the patients and were related to a combination of radiation therapy, surgical treatment, and HSCT in addition to chemotherapy (11, 14, 15). In Japan, Shimazaki et al. reported that 56 (81.1%) of 69 patients with CCS had endocrine late effects, 13 (18.8%) experienced GH deficiency, 14 (20.2%) experienced hypothyroidism, and 25 (36.2%) experienced hypogonadism (11). Miyoshi et al. reported that 82 (67%) of 122 CCS patients had endocrine late effects, 39 (32%) had growth disorders, 26 (21%) had thyroid dysfunction, and 60 (49%) had gonadal dysfunction (14). In addition, obesity was reported in 20 (16%) patients, adrenal dysfunction in 9 (7%), and central diabetes insipidus in 11 (9%) patients in this report. These reports do not necessarily indicate that complications occur with the same frequency in infants with ALL because of the wide range of primary diseases, age at diagnosis, timing of treatment, and the effects of different treatments. Long-term complications, including gonadal dysfunction, are often unknown because of the small number of survivors of infantile ALL. In this study, we aimed to investigate the actual status of endocrine late complications in infantile ALL survivors and discuss the points to be considered during follow-up.

In a previous study reported in 2007 regarding infantile ALL survivors with KMT2A gene rearrangement, 23 (63.8%) of 36 transplanted patients treated with the JILSG MLL96/98 protocol had short stature and 5 (13.8%) had hypothyroidism (2). In this previous study, 22 transplant recipients were pretreated with TBI, and the median height SDS at the time of the study was –3.07SD in the TBI group, which was clearly impaired, compared with the non-TBI group (–1.72SD) (2). In the present study, 2 of 6 patients had short stature and 2 had hypothyroidism. One patient (Case 6), who had undergone two transplants and one TBI, showed significantly short stature and thinness, which may have been related to a history of radiation therapy.

Gonadal function was not assessed in the aforementioned report because the oldest child at the time of the survey was 10 yr old and had not yet reached the age of puberty (2). We evaluated gonadal function in six infantile ALL survivors treated at our hospital based on CED, which was previously reported. Among the six cases, all three girls presented with primary hypogonadism, and one boy (Case 6) was also suggested to have impaired spermatogenesis and testosterone deficiency. Although in a small number of cases, the results suggest that hypogonadism may be more frequent than short stature or hypothyroidism in survivors of infantile ALL. In previous reports on the relationship between CED and the development of hypogonadism, girls with a CED of > 8 g/m² presented with ovarian dysfunction and required gonadal hormone replacement therapy (4, 6). In boys, testosterone secretion decreases with a CED of > 20 g/m², and testosterone replacement therapy was required (4). However, other reports have shown a decrease in sperm count and sperm concentration in boys exposed to a CED of > 4 g/m², suggesting that male infertility may be a problem after reproductive age, even if puberty development is not impaired (5). In this study, primary hypogonadism was suggested in all three girls with a CED of > 8 g/m² and in the boy in Case 6 with a CED of > 20 g/m², findings that are consistent with previously reported risk factors. Furthermore, the boy in Case 5, who had reached puberty, also had a CED of > 4 g/m² and should be followed carefully for future spermatogenesis. In boys with CCS, it is difficult to assess the onset of puberty by testicular volume alone because testicular size is reduced in cases of impaired seminiferous tubule function (12). It is important to comprehensively evaluate the onset of puberty based on factors such as the growth spurt and bone age as well as testicular volume.

In infantile ALL, the treatment protocol until 2009 (JPLSG MLL03) required transplantation in all patients with the KMT2A gene rearrangement. However, in recent years, it has become clear that some patients with gene rearrangements had a positive response to chemotherapy alone without transplantation (16). Although large amounts of alkylating agents have been used as pre-transplant therapy in prior transplant cases, attempts have recently been made to reduce the risk of late complications. The latest protocol (JPLSG MLL10) categorizes the risk and recommends that transplantation should not be performed in patients with genetic reconstruction who are at least six months old and do not have central nervous system involvement (intermediate group) (16). Therefore, the number of patients eligible for transplantation has decreased. However, when compared with the JPLSG ALL-B12 protocol, which is commonly used in children older than 1 yr, the non-transplant group (standard risk group) was 51% and the high-risk group requiring transplantation was 17% in ALL-B12, whereas the non-transplant group (intermediate-risk group) was 21% and the transplant group (high-risk group) was 62% in MLL10 (16, 17). In addition, the non-transplant group in ALL-B12 for children older than 1 yr has a CED of 3.0 g/m², and the transplant group has a CED of 12 g/m², whereas the non-transplant group in MLL10 for infantile ALL has a CED of 7.0 g/m², and the transplant group has a CED of 8.6 g/m². Thus, since transplantation cases are still common and CED values are still high even without transplantation in infantile ALL, hypogonadism may be more frequent in endocrine late complications than in other childhood cancers.

In the present study, one patient had type 2 DM and hyperlipidemia without obesity. Insulin resistance without obesity and abnormal lipid metabolism have been reported as late complications in patients with CCS. Bizzarri et al. reported that a history of TBI and a significant length of time since HSCT were risk factors for the acquisition of insulin resistance at a young age (18). In a study of DM risk among 8,599 CCS, an increased risk of DM was associated with TBI, abdominal irradiation, alkylating agents, and younger age at diagnosis (19). Friedman et al. have reported that TBI is a risk factor for metabolic disorders without obesity (20). In some cases, patients who underwent TBI for HSCT developed partial lipodystrophy, a condition similar to metabolic syndrome, without obesity (9, 21, 22). Most patients diagnosed with HSCT-associated partial lipodystrophy display a similar pattern of body fat distribution with reduced subcutaneous fat, especially at the extremities, and preserved or even enlarged fat deposits in the cheeks, neck, and abdomen. Increased visceral fat deposition associated with fatty liver disease has also been reported in most cases (9). While TBI is considered a main risk factor for partial lipodystrophy, chronic GVHD, chemotherapy, and immunosuppressants may also be involved in glucose and lipid metabolism disorders. In our study, while few children had received TBI, the age at which they received therapy for ALL was younger and the time since treatment was longer. In particular, Case 3 presented with abnormalities in glucose and lipid metabolism and partial lipodystrophy despite not having undergone TBI. Therefore, it is important to consider that infantile ALL patients may show abnormalities in glucose and lipid metabolism by the time they reach adulthood.

Our study has some limitations. This is a small, single-center, retrospective study, and some data are inadequate. However, we were able to obtain valuable data, including long-term complications, from a small number of survivors of infantile ALL. More data are required in the future.

Conclusion

A high proportion of infants require transplantation in the treatment of infantile KMT2A-r ALL, and the dose of alkylating agents is high. Regarding endocrine late effects, the risk of developing primary hypogonadism, with addition to growth disorders and hypothyroidism, should be carefully monitored. Furthermore, CED is useful in predicting the development of primary hypogonadism.

Conflict of interests

The authors declare no conflicts of interest.