Pulsatile growth hormone secretion in children with acute lymphoblastic leukemia after 1800 cGy cranial radiation

Pulsatile growth hormone secretion in children with acute lymphoblastic leukemia after 1800 cGy cranial radiation

0360~3016/H $3.00 + .OO Copyright 0 1988 Pergamon Press plc ht. J. Radarion OncologyBio/. Phys., Vol. 15, pp. 100-1006 printed in the U.S.A. All righ...

576KB Sizes 13 Downloads 46 Views

0360~3016/H $3.00 + .OO Copyright 0 1988 Pergamon Press plc

ht. J. Radarion OncologyBio/. Phys., Vol. 15, pp. 100-1006 printed in the U.S.A. All rights reserved.

0 Brief Communication PULSATILE GROWTH HORMONE SECRETION IN CHILDREN WITH ACUTE LYMPHOBLASTIC LEUKEMIA AFTER 1800 cGy CRANIAL RADIATION JULIE BLATT,

M.D.,*

PETER LEE, M.D.,

PH.D.,? AND DAVID FINEGOLD,

JACQUELINE M.D.

SUTTNER,

R.N.?

Department of Pediatrics, Children’s Hospital of Pittsburgh, Pittsburgh, PA 15213 The relationship between intensity of central nervous system preventive therapy and the development of hypothalamic pituitary dysfunction is unclear in patients with acute lymphoblastic leukemia. In a previous report, we demonstrated uniform decreases in spontaneous secretion of growth hormone following 2400 cGy whole brain radiation. In this study, we measured basal growth hormone levels every 20 minutes over 24 hr in five survivors of childhood acute lymphoblastic leukemia treated with 1800 cGy cranial radiation. Four of the patients had been off therapy 2&4& years. Growth hormone secretion in these patients, as indicated by mean growth hormone concentration, pulse amplitude and frequency, was clearly greater than that seen following 2400 cGy and appeared to be normal compared with sex- and Tanner stage-matched literature controls. However, serial growth measurements showed significant decreases in height percentiles in two of these children. The fifth patient, who had already approached her adult height at the time of diagnosis, had been off therapy only 1 year and had a mean growth hormone level intermediate between those of normal controls and previously reported children treated with 2400 cGy. These data suggest (a) that the effect of radiation therapy on spontaneous pulsatile growth hormone secretion may be dose related, and (b) that short stature in a given patient may not be indicative of subnormal basal growth hormone levels. Further longitudinal investigation may clarify whether early transient changes in GH secretion occur that may normalize over time. Acute lymphocytic leukemia, Central nervous system radiation, Growth hormone.

means of assessing treatment-related neuroendocrine damage.’ In the present study, we measured basal growth hormone levels over 24 hr in a group of patients who had received 1800 cGy cranial radiation.

INTRODUCIION

Increasing numbers of children with acute lymphoblastic leukemia (ALL) are being cured of their disease. Where therapeutic results have been maximized, attempts are being made to decrease the intensity of therapy to minimize its late sequelae. One trend has been toward the reduction in dosage, or even elimination, of cranial irradiation in the preventive management of central nervous system (CNS) leukemia. Whether this strategy will be successful is as yet unclear. In a previous study of long-term survivors of childhood ALL, we reported that 2400 cGy cranial radiation blunted spontaneous pulsatile secretion of growth hormone, and suggested that evaluation of spontaneous pulsatile growth hormone secretion may provide a sensitive

METHODS

Patients Five patients with a history of childhood ALL who are being followed at the Children’s Hospital of Pittsburgh were studied after informed consent had been obtained. Patient characteristics are summarized in Table 1. To optimize comparability with respect to therapy with children studied previously,* patients were selected who had been treated according to Childrens Cancer Study Group Protocols 16 1 or 162,3 which included prednisone, vin-

* Division of Hematology-Oncology. t Division of Endocrinology. Supported in Part by Research Grant #2MO 1RR00084 from the National Institutes of Health. Reprint requests to: Julie Blatt, M.D., Children’s Hospital of Pittsburgh, One Children’s Place, 3705 Fifth Avenue at DeSoto Street, Pittsburgh, PA 15213-3417. Acknowledgements-We would like to thank Drs. V. Albo, W.

Prin, S. Orlando, and M. Wollman for permission to include their patients in this study, the nursing staff of the Clinical Research Unit of Children’s Hospital of Pittsburgh for processing samples for growth hormone determinations, Ms. D. Cleary for performing growth hormone assays and Ms. C. White for performing somatomedin C assays. Accepted for publication 2 1 April 1988.

1001

1002

1. J. Radiation Oncology 0 Biology 0 Physics

October 1988, Volume 15, Number 4

TIME

TIME

=

E .

s

1 8

IO

5

I 6

TIME

Fig. 1. Spontaneous growth hormone secretion in patients no. 1 (a), 2 (b), 3 (c), 4 (d), 5 (e).

cristine, 1-asparaginase, &mercaptopurine, and methotrexate. One patient (#5, Table 1) also received monthly doses of intravenous cyclophosphamide for 2 years as part of maintenance therapy. In addition, in contrast to our previous study in which patients had received 2400 cGy in 200 cGy fractions over 2i weeks, each of the pa-

tients received CNS preventive therapy with 1800 cGy cranial radiation given in 180 cGy fractions over 2 weeks. The present patient cohort also received intrathecal methotrexate beginning during induction therapy and continuing through radiation into maintenance therauy. In contrast, the 2400 cGy group received the

1003

Growth hormone and acute leukemia 0 J.BLATT etal

’ IO

L

I

I

I

I

1

12

2

4

6

Sp

I

I

12

IO

I

I

1

I

2

4

6

SA

TIME

= .E z

I5

W g

IO

g I E s

5

8

2

. 9

;o

I

I

;

;

;

;

;’

;

;o

Ii

1;

i

;

;

;

;

TIME Fig. 1. (Co&)

first intrathecal methotrexate concurrent with radiation. Median time from completion of radiation therapy was 54years (range 3&-7& years). Median time from completion of all therapy was 3$ years (range l-4& years). Evaluation of study participants Patients were admitted to the Clinical Research Unit, Children’s Hospital of Pittsburgh. Growth charts were compiled by review of hospital records, using standard growth forms. Heights had been recorded at diagnosis, monthly during therapy, every 3 to 4 months for the first year off therapy, and yearly thereafter. At the time of the present evaluation, height was measured three times by

stadiometry and the average recorded. To evaluate spontaneous pulsatile secretion of growth hormone, 1 ml heparinized blood samples were obtained every 20 minutes for 24 hr from an indwelling heparin lock placed in a forearm vein. In addition, a single plasma specimen was obtained at the time of initial sampling for measurement of somatomedin C levels. Each time a blood sample was removed, it was noted whether the patient was asleep or awake so that growth hormone levels could be correlated with sleep. An effort was made to avoid stress and encourage normal activity throughout the hospital stay. Wrist radiography for bone age determination was performed on several patients.

1004

I. J. Radiation Oncology 0 Biology 0 Physics

October 1988, Volume 15, Number 4

Table 1. Characteristics of study participants Patients

Age (years)*

Sex

Tanner stage

:

3 5/12-8 3/12 8112-10 2112 2 1l/12-8 6/12 14 8/12-22 12 l/12-15 8/12

F M

3 4 5

F M F

Height (%)

Mean GH @g/ml)

GH peaks? 8AM-8PM

GH peaks? 8PM-8AM

SMC (u/ml)

I

50150

2.28 4.96

I V V

65130 90150 80180 90190

2.06 2.24 1.80

22(16-17, 17) (5-7,6) 2(9-l 1, 10) 0 0

53 (6-l 3,9) (9-13, 13) 4 (6-9,8) 2 (28-9, 18)

0.50 0.96 0.52 1.61 1.08

l(18)

* At diagnosis of ALL - at time of present study. t Number of peaks (range of peak amplitudes, median peak amplitude rounded to nearest whole number).

Measurement of hormone levels For each subject, growth hormone samples were processed at one time to avoid interassay variability. Growth hormone and somatomedin C were assayed by specific radioimmunoassay. 5-gMean growth hormone concentration over 24 hr was calculated using the Pulsar-Peak Identification Algorithm Program.6 Growth hormone secretion was analyzed by G criteria such that the likelihood of detecting a falsely positive pulse was less than 2%. The frequency and amplitude of pulses of growth hormone secretion were noted and analyzed in terms of daytime (8 AM to 8 PM) and nocturnal (8 PM to 8 AM) secretion. For purposes of comparison, data from the prior study of patients who had received 2400 cGy cranial radiation* were re-analyzed using the same abovenoted Algorithm. In addition, reported data from endocrinologically normal sex- and Tanner stage-matched children2,12 were available for comparison. RESULTS Patient characteristics are summarized in Table 1. Patients # l-4 were off all therapy for ‘25 years-4h years and were 4+-7& years from completion of radiation therapy. The median decrease in height percentiles from the time of diagnosis to the present was 18 (range O-40%). Two patients had decreases of 35 and 40 percentiles. Patients #4 and #5 had already approached their adult heights prior to diagnosis and both of them had normal adult bone ages. In all five patients growth hormone output, as measured by the frequency and amplitude of growth hormone pulses, showed normal diurnal variation and was greater at night than during the day (Fig. 1). Mean 24 hr growth hormone levels ranged from 1.80 to 4.96 ng/ml (2.66 + 1.30, SD.), which is comparable to the level of 2.9 f 0.02 ng/ml reported previously in a population of normal prepubertal children (values for pubertal children were statistically comparable).‘2 In contrast, data re-analyzed from our past study of children treated with 2400 cGy showed a mean growth hormone level of .4 14 + 0.40 ng/ml (p 5 0.001, compared with the 1800 cGy group). Of the patients in the present study, the lowest

mean 24 hr growth hormone level was measured in the patient who had been off therapy for the shortest period of time. Among the other four patients, the median number of pulses r5 ng/ml over a 24 hr period was 5 1 (range 2-7) compared with 1 (range O-3) in our previous study with children treated with 2400 cGy cohort (p = 0.05). The median pulse amplitude in the 1800 cGy cohort was 10 ng/ml (range 5-28 ng/ml) compared with 3 ng/ml (range 0- 13 ng/ml, p = 0.0 1) in our patients treated with 2400 cGy. The fifth patient in the present study produced only a single nocturnal pulse of growth hormone ( 18 @ml). In our previous report,2 among normal subjects who were of comparable sex and Tanner stage to that of the present patients, a median of two daytime and four nocturnal peaks were noted. As determined by mean growth hormone level, pulse number and amplitude, this patient’s growth hormone output would be considered to be intermediate between what was seen in normal individuals and those treated with 2400 cGy. Somatomedin C levels in all patients were within normal limits. DISCUSSION Children treated for ALL with central nervous system preventive therapy, including intrathecal chemotherapy and cranial radiation, may experience a number of delayed neuroendocrine side effects. Among these is a decrease in growth velocity with resulting short stature in some patients. The presence or absence of abnormal growth hormone responses to a variety of provocative stimuli has not always correlated with the finding of short stature in particular individuals.7*9s1“‘3~‘5Thus, there has not been a reproducible biochemical marker for, or predictor of, short stature. We and others have reported that basal growth hormone levels measured every 20 minutes over 24 hr are uniformly blunted in long-term survivors of childhood ALL whose regimens have included 2400 cGy to the whole brain.2~’’Although blunted spontaneous release of growth hormone in these studies was not always associated with short stature, we postulated that this abnonnality might be a marker of neuroendocrine damage and an early change preceding other abnormalities which would

Growth hormone and acute leukemia 0 J. BLATT et al.

lead to short stature. Because only patients treated with 2400 cGy were studied, the relationship between intensity of CNS preventive therapy and the development of abnormalities in spontaneous pulsatile growth hormone secretion was not clear. In the present study, we looked at five survivors of ALL who received CNS preventive therapy with only 1800 cGy cranial radiation, a dosage no longer used at our institution in patients receiving the minimal therapy comparable to our past study. The patterns of GH secretory pulses stand in sharp contrast to the previously reported results following 2400 cGy. However, they appeared to be comparable to those of age-matched normal individuals: for example, mean growth hormone levels for our patients treated with 1800 cGy, our previously reported patients treated with 2400 cGy, and untreated controls were 2.66, 0.414, and 2.9 ng/ml, respectively. One of our five patients (the one who had been off therapy the shortest period of time) had a 24 hr level of 1.8 ng/ml which falls more than 2 standard deviations below that measured in a group of normal Tanner stage matched girls. I2 However, she had one spontaneous growth hormone peak of 18 ng/ml, more than 2 standard deviations above that seen in our prior study in patients treated with the higher dose of radiation therapy. Moreover, 18 ng/ml also is well above the 5 ng/ml level commonly used for the clinical diagnosis of growth hormone deficiency. Somatomedin C levels were normal in all five patients, whereas they had been low in two out of seven leukemic children following 2400 cGy from our prior report.’ Chemotherapy was similar in all the irradiated children in both of these studies. Thus, the abnormalities in growth hormone experienced by the children in the 2400 cGy group apparently relate to radiation therapy. Therefore, based on measurements of spontaneous pulsatile growth hormone secretion several years after cranial radiation, 1800 cGy appears to have less neuroendocrine toxicity than 2400 cGy. A recent retrospective study has noted a possible protective effect of intrathecal methotrexate on long-term neuropsychologic function

1005

when given in advance of cranial radiation.’ Thus, it is possible that patients in the present study, who received intrathecal methotrexate during induction therapy, might have had less toxicity on that basis than those whose first intrathecal methotrexate was given concurrent with radiation. Interestingly, despite apparently normal growth hormone and somatomedin C production, three patients, all of whom were treated before puberty, had significant falloff in height percentiles. Similarly, in a larger group of patients who had not undergone hormone studies, we have found that 1800 cGy and 2400 cGy have comparable negative effects on height,14 a finding which is supported by previous reports.‘oV’6Thus, the physiologic significance of the normal basal growth hormone pulses in these patients in unclear. It may be that they had abnormal secretion of growth hormone earlier in their postradiation course which has normalized over time. This hypothesis is suggested by the results from our fifth patient. At the time of her evaluation, that child had been off all therapy at least 2; years. She produced only a single growth hormone pulse greater than 5 ng/ml. Reversible abnormalities of spontaneous pulsatile growth hormone secretion following even 2400 cGy have been reported.4 Eighteen hundred rad appears to have comparable efficacy to 2400 cGy whole brain radiation in preventing the development of CNS leukemia.3 Because it was anticipated that the lower radiotherapy dose would have less toxicity, it is disappointing that the adverse effects on attainment of adult heights of 1800 cGy and 2400 cGy are comparable. Nonetheless, the differential effect on delayed measurements of spontaneous pulsatile growth hormone secretion raise the possibility that further reduction of radiation dosage may spare growth hormone secretion (and presumably growth) altogether. Evaluation of additional patients with serial measurements of spontaneous pulsatile growth hormone secretion, starting within several months of radiation therapy as well as with measurement of bone ages, should be informative in addressing this issue.

REFERENCES 1. Balsom, B., Bleyer, A., Robison, L., Heyn, R., Meadows, A., Sitarz, A., Miller, D., Leikin, S., Sathers, S., Ortega, J., Nesbit, M., Hammond, D.: Cranial irradiation in young children: Can intellectual deficits be prevented with pretreatment methotrexate. Proc. Am. Sot. Clin. Oncol. 6: 10, 1987. 2. Blatt, J., Bercu, B.B., Gillin, J.C., Mendelson, W.B., Poplack, D.G.: Reduced pulsatile growth hormone secretion in children after therapy for acute lymphoblastic leukemia. J. Pediut. 104: 182-186, 1984. 3. Coccia, P.F., Bleyer, W.A., Siegel, S.E., Lukens, J.N., Gross, S., Miller, D.R., Littman, P.F., Sather, J.M., Hammond, D.: Development and preliminary findings of Childrens Cancer Study Group protocols (CCG- 16 1, 162, 163) for low, average, and high risk acute lymphoblastic leukemia in children. In Leukemia Research, Advances in Cell

4.

5.

6.

7.

Biology and Research, Murphy, S., Gilbert, J.R. (Eds.). North Holland, Elsevier Publishing. 1983, pp. 24 l-250. Dacou-Vaitekakis, C., Zypolyta, A., Haidas St. Constantinidis, M., Papavasikiou, C., Zannos-Mariolea, L.: Irradiation of the head: Immediate effect on growth hormone secretion in children. J. Clin. Endocrinol. Met. 4: 79 l-794, 1977. Furlanetto, R.W., Underwood, L.E., Van Wyke, J.J., D’Ercole, A.J.: Estimation of somatomedin C levels in normals and patients with pituitary disease by radioimmunoassay. J. Clin. Invest. 60: 648-657, 1977. Merriam, G.R., Wachter, K.W.: Algorithm for the study of episodic hormone secretion. Am. J. Phys. 243: E310318, 1982. Muhlendahl, K.E., Gadner, H., Riehm, H., Helge, H., Weber, B., Mutterhess, R.: Endocrin function after anti-neo-

1006

I.J. Radiation Oncology 0 Biology 0 Physics

plastic therapy in 22 children with acute lymphoblastic leukemia. Helv. Paediatr. Acta 31: 463, 1976. 8. Odell, W.O., Rayford, P.L., Ross, G.T.: Simple partially automated method for radioimmunoassay of human thyroid stimulation, growth, leuteinizing and follicle stimulating hormone. J. Lab. Clin. Med. 70: 978, 1967. 9. Oliff, A., Bode, U., Bercu, B.B., Dichiro, D., Graves, V., Poplack, D.G.: Hypothalamic pituitary dysfunction following CNS prophylaxis in acute lymphocytic leukemia: Correlation with CT scan abnormalities. Med. Pediat. Oncol. 7: 141-151, 1979. 10. Robison, L.L., Nesbit, M.E., Sather, H.N., Meadows, A.T., Ortega, J.A., Hammond, G.D.: Height of children successfully treated for acute lymphoblastic leukemia: A report from the late effects study committee of Children’s Cancer Study Group. Med. Pediat. Oncol. 13: 14-2 1, 1985. 11 Romshe, CA., Zipf, W.B., Miser, J., Sotos, J.F., Newton, W.A.: Evaluation of growth hormone release and human growth hormone treatment in children with cranial irradiation-associated short stature. J. Pediat. 104: 177-18 1, 1984.

October 1988, Volume 15, Number 4 12. Ross, J.L., Long, L.M., Loriaux,

D., Cutler, G.B. Jr.: Growth hormone secretory dynamics in Turner syndrome. J. Pediat. 106: 202-206, 1986.

13. Shalet, A.M., Beardwell, C.G., Morris-Jones, P.H., Pearson, D.: Growth hormone deficiency after treatment of acute leukemia in children. Arch. Dis. Child. 51:489-493, 1976. 14. Starcewskj, P.J., Lee, P.A., Blatt, J., Finegold, D., Brown, D.: Comparable effects of 1800 and 2400 rad cranial irradiation on height and weight in children treated for acute lymphocytic leukemia. Am. J. Dis. Child. 141: 550, 1987. 15. Swift, P.G.F., Kearney, P.J., Dalton, R.G., Bullimore, J.A., Mott, M.G., Savage, D.C.L.: Growth and hormonal status of children treated for acute lymphoblastic leukemia. Arch. Dis. Child. 53: 890-894, 1978. 16. Wells, R.J., Foster, M.B., D’Ercole, A.J., McMillan, C. W.: The impact of cranial irradiation on the growth of children with acute lymphocytic leukemia. Am. J. Dis. Child. 137: 37-39, 1983.