Growth Hormone Use in Pediatric Growth Hormone Deficiency ...€¦ · Evaluation for growth hormone (GH) deficiency in a child who is short (ie, whose height is more than 2 standard

VOL. 6, NO. 15, SUP. THE AMERICAN JOURNAL OF MANAGED CARE S805

Growth Hormone Use in Pediatric Growth HormoneDeficiency and Other Pediatric Growth Disorders

Based on presentations by Ron Rosenfeld, MD; David B. Allen, MD; Margaret H. MacGillivray, MD;Craig Alter, MD; Paul Saenger, MD; Henry Anhalt, DO; Ray Hintz, MD; and Harvey P. Katz, MD

. . . SYMPOSIUM PROCEEDINGS . . .

Presentation Summary The diagnosis and management

of growth disorders in children,particularly disorders that respondto therapy with growth hormone(GH), raise challenging clinical andeconomic issues. Several suchissues are presented in the followingarticle in which Dr. Ron Rosenfeldexamines the evaluation and diag-nosis of the child with short stature;Dr. David B. Allen discusses theanabolic and metabolic indicationsfor GH treatment in children; Dr.Margaret H. MacGillivray reviewsGH dosing, height outcomes, and

follow up; and Dr. Craig Alter pres-ents the payer’s perspective on thediagnosis and treatment of pedi-atric GH deficiency.

In addressing the use of GH inother pediatric populations, Dr.Paul Saenger focuses on Turnersyndrome, Dr. Henry Anhalt onchronic renal insufficiency ofchildhood, and Dr. Ray Hintz onidiopathic short stature. Dr. HarveyP. Katz presents one managed careorganization’s policy and imple-mentation plan that is used toguide decisions regarding coveragefor GH treatment.

Payer Highlights

● Concrete policies/guidelines on the use of GH that are authoredby endocrinologists would assist managed care plans in makingcoverage decisions that are consistent with the endocrine society.

● Guidelines must allow for flexibility to determine coverage on acase-by-case basis, with regard to new indications and technology.

● Case review process must be refined: o Payers require detailed patient history from clinicians to deter-

mine whether coverage for GH therapy should be approved. o A review committee comprising an endocrinologist, a medical

director, and a payer could examine efficacy and safety oftreatment in each case, and whether the decision to cover isfair and ethically sound as well as update guidelines periodi-cally to reflect developments in the endocrine community.

Growth in humans is anexceedingly complex biologicprocess, and growth disorders,

which are most apparent in childhood,are equally complex. The diagnosisand management of such disorderspose challenging clinical and econom-ic issues; several of which are exploredin the following presentations.

Evaluation and Diagnosis of the Child With Short Stature

When evaluating a child with shortstature, the pediatric endocrinologistoften integrates multiple clinical andlaboratory data, as is done with anyother endocrine diagnosis, ratherthan relying only on the results of aparticular biochemical assay.

Evaluation for growth hormone(GH) deficiency in a child who isshort (ie, whose height is more than 2standard deviations [SDs] below themean) should not be initiated untilother causes of growth failure, such ashypothyroidism, chronic systemicdisease, Turner syndrome, or skeletaldisorders, have been excluded. Acomprehensive clinical and auxologicassessment, combined with a bone-age X ray and biochemical tests of theGH/insulin-like growth factor (IGF)axis, is then warranted. The auxolog-ic assessment consists primarily of acareful evaluation of the child’sgrowth rate and velocity and familyhistory of growth and height patterns.

GH, a major peptide hormone, issecreted by the anterior pituitary. Itbinds to a transmembrane receptorand eventually leads to the produc-tion of at least 3 GH-dependent pep-tides: IGF-1, IGF binding protein-3(IGFBP-3), and the acid-labile sub-unit. These 3 peptides combine toform the major circulating form ofIGF, which is delivered to the targetcell, binds to an IGF-1 receptor, andstimulates skeletal growth.

Because GH is normally secretedin a pulsatile manner, with very lowlevels seen during the day and 6 or 7spontaneous spikes most commonly

occurring during deep sleep, clini-cians cannot rely on a random bloodsample for GH level measurement.Instead, endocrinologists have stimu-lated the pituitary gland pharmaco-logically (eg, with insulin) or physio-logically (via exercise or sleep) tosecrete GH to observe how well thegland responds. However, even innormal individuals rendered hypo-glycemic by intravenous insulin, theincrease in GH levels in response tohypoglycemia varies considerably,from an arbitrarily defined, andassay-dependent, normal level of atleast 10 ng/mL to levels that are sev-eral times higher.

The arbitrary nature of suchprovocative testing is illustrated by astudy in which spontaneous GHsecretion was measured via anindwelling intravenous line every 20minutes over a 24-hour period in 2normal short children (normal on thebasis of provocative GH levels ofgreater than 10 ng/mL) and 2 chil-dren diagnosed with GH deficiency(GH deficient on the basis of provoca-tive GH levels less than 10 ng/mL).Although 1 GH-deficient child had lit-tle measurable GH, the other exhibit-ed 2 spontaneous peaks of 20 ng/mLand a GH profile that was indistin-guishable from those of the 2 normalshort children.

The tests rely on an arbitrary defi-nition of what constitutes a normal orsubnormal response, the assays usedto measure response are not stan-dardized and have poor reproducibili-ty, and the response to testing is agedependent in children and adults. Inaddition, test results are affected tosome extent by sex steroids. Thus,criteria used to diagnose prepubertaland pubertal children must be recon-sidered. Because such tests lackbimodal distribution, they are limitedin their capacity to identify partialGH deficiency. Those tests also areuncomfortable, expensive, and attimes involve an element of risk tothe patient.

S806 THE AMERICAN JOURNAL OF MANAGED CARE SEPTEMBER 2000



An alternative approach is to con-sider the 3 plasma components of theIGF system that are GH dependent(IGF-1, IGFBP-3, and the acid-labilesubunit) in addition to GH measure-ments. However, measurement of theIGF components also has limitations.For example, IGF-1 levels in childrenare markedly age dependent; verylow levels are observed in youngchildren. Moreover, comparisonsbetween the results of IGF-1 assaysand GH provocative tests are not100% concordant. Although mostchildren who have failed provocativetesting also have a low IGF-1 level,some children with low GH levelshave a normal IGF-1 level. Similarly,many children with idiopathic shortstature have a normal GH level but alow IGF-1 value. As is the case withGH levels, there is no bimodal distri-bution of IGF-1 concentrations andno clean separation of abnormally lowlevels from lower limits of normal.Although more normative IGF-1 datawould be of value, it is likely that cli-nicians are dealing with a continuumof GH-IGF secretion, and sharp dis-crimination between GH deficiencyand borderline GH-IGF productionmay not be possible.

This is also the case with theIGFBP-3 level. Although many chil-dren with classic GH deficiencyexhibit very low levels of IGFBP-3,some have a normal IGFBP-3 value.Again, there is not 100% concordancein the results of IGFBP-3 assays andGH tests.

The inability to reconcile theresults of GH and IGF tests may berelated more to the inadequacies ofGH testing than to the relevance ofIGF measurements. At present, ampledata support the primacy of IGF deter-minations in truly defining theendocrine milieu of a pediatric patient.However, it is probably unrealistic toexpect—or to have ever expected inthe past—that a biological process ascomplex as growth can be reduced to 1or 2 simple diagnostic tests.

At this time, the diagnosis of GHdeficiency in children with shortstature may need to be based onauxology and supported by eitherprovocative GH testing or carefuldeterminations of the IGF system.That approach may result in someoverdiagnosis and overtreatment,both of which are preferable tofailing to diagnose or to treatappropriate children during a win-dow of time that will allow them togrow.

Anabolic and MetabolicIndications for GH Treatment in Children

In addition to promoting lineargrowth, GH has considerable andfavorable physiologic effects on adi-pose tissue, bone metabolism, andmuscle accretion. During childhood,GH stimulates the proliferation ofadipose precursor cells, restrictstheir differentiation into matureadipocytes, stimulates hormone-sen-sitive lipase (the rate-limiting step forlipolysis), and limits deposition of fatin the abdominal visceral depot.Normal secretion of GH also pro-motes accretion of lean body mass,increases muscle weight and nitrogenretention (both directly and indirect-ly via IGF-1), and appears to influ-ence normal muscle fiber distributionbetween type 1 and type 2 musclefibers. Moreover, because GH andIGF-1 receptors are expressed in var-ious sites in bone, GH directly stimu-lates osteoblast and osteoclast differ-entiation and promotes the accretionof bone mass during childhood andadolescence. Consequently, childrenwith severe GH deficiency demon-strate reduced linear growth andreduced lean body mass, increasedbody fat (particularly as a dispropor-tionate deposition of visceral andtruncal fat), subnormal bone density,and a tendency to develop lipidabnormalities.

Traditionally, the indications forGH therapy in childhood have been

. . . PEDIATRIC GROWTH HORMONE DEFICIENCY AND OTHER PEDIATRIC GROWTH DISORDERS . . .


judged by growth impairment and tar-get height goals. Therefore, normallygrowing GH deficient children, suchas those who have undergone cranio-pharyngioma removal, often do notreceive GH therapy. However,increased knowledge about the differ-ent physiologic effects of GH hasraised several issues, primarily that ofweighing the benefit of treatmentagainst its risks, its costs, and the pos-sible treatment-related morbidity.Although the value of GH therapy inadults with GH deficiency has beenrecognized and such treatment hasbeen approved because of its benefi-cial effects on body composition andlipid profiles, this is not yet the casein children with GH deficiency.However, the principles used to deter-mine the value of treatment for adultswith GH deficiency should be appliedto children as well.

Body Composition. Children under-going surgery to remove a cranio-pharyngioma demonstrate a well-known phenomenon of growth withoutGH. Often, they receive no GH therapyafter surgery because there is noheight-related indication for suchtherapy. However, because manysuch children exhibit markedlyabnormal body composition (mostnotably, truncal obesity) despite nor-mal or near-normal height for theirage, it is highly likely that the lack ofGH is responsible for the increase infat mass and the reduction in leanbody mass.

Body composition studies are per-formed to determine how to make thetransition from GH therapy in child-hood to possible GH therapy inadulthood. However, the effects of GHon body composition in GH deficientchildren have not often been studiedin a prospective fashion. One suchstudy1 of a mixed population of GHdeficient children with increased fatmass, markedly reduced lean bodymass, and some decline in total bonemineral content at baseline indicated

that there was a statistically signifi-cant reduction in fat mass and anincrease in lean body mass after 6months of GH therapy and a signifi-cant improvement in bone mineralcontent after 12 months of treatment.Favorable changes in body composi-tion continued with increasing dura-tion of therapy, as did improvementin the lipid profiles, which wereabnormal at baseline.

Another study of 65 prepubertalchildren receiving GH and 35 individ-uals who had completed GH therapy6 months earlier found that treatmentwas associated with a remarkabledecline in body fat.2 The girls studieddemonstrated a normal age-relatedincrease in fat mass over time, but theboys retained the reduction in thepercentage of body fat. However,when GH was discontinued, thereturn of abnormal body compositionwas more pronounced in boys than ingirls, as was the return of abnormalcholesterol profiles.

Prader-Willi Syndrome. The valueof GH therapy to correct abnormali-ties of body composition that are sep-arate from the issue of height in chil-dren is well illustrated in Prader-Willisyndrome, an inherited functionaldeletion of chromosome 15 thataffects approximately 1 in 15,000children. The syndrome is character-ized by obesity, hyperphagia, hypoto-nia, hypogonadism, hypothalamicdysfunction, and short stature,although the latter is not as severe asthat seen in individuals with GHdeficiency.

Controversy exists about whetherchildren with the syndrome are trulyGH deficient. It is very difficult to testthose children for hormone deficiencybecause their obesity suppresses GHproduction. However, increasing evi-dence indicates that those childrenare GH deficient, because they stopgrowing when they are placed on acalorically restricted weight reduc-tion diet. Apparently, overnourish-



ment is required for growth in suchchildren.

In a study of 54 children withPrader-Willi syndrome, Carrel andassociates evaluated the effects of GHtreatment on body composition,strength and agility, and respiratorymuscle strength. Their 24-month fol-low-up results indicate significantincreases in linear growth rates, leanbody mass, and bone mineral density;a decrease in fat mass; improved agili-ty and physical functioning; and noprogression of scoliosis or glucoseintolerance.3

For such children and their par-ents, the value of GH therapy is not inthe increase in height but rather inimproved physical function and abili-ty to do things that could not be donebefore treatment, such as play half agame of soccer, climb on the schoolbus unassisted, or ride a bike. Froman ethical point of view, it is theresponsiveness to GH therapy and theimproved functional outcome thatshould determine entitlement totreatment not the diagnosis of GHdeficiency.

GH Dose, Height Outcomes, and Follow Up

Before 1985, the adult height ofchildren with GH deficiency whowere treated with pituitary GHranged from -4.7 to -2.0 SD. Since1985, when recombinant GHbecame available, the adult height ofchildren treated with recombinantGH has ranged from -1.5 to -0.7 SD.The marked improvement in adultheight has been attributed to higherdoses of recombinant GH (2 to 3times higher than those used withpituitary GH), standardization ofdose by weight, earlier chronologicand bone age at the start of treat-ment, longer duration of therapy,correction of height deficit beforethe onset of puberty, the frequencyof hormone administration, parentalheights, gender, and compliancewith therapy.

The GH dose and the duration oftherapy are critical to favorableheight outcomes. When GH deficientchildren were treated with pituitaryGH, they were often given a fixed doseof 2 units, regardless of body size.Now, children treated with recombi-nant GH are given a higher dose thatis standardized to weight: 0.2 to 0.3mg/kg/wk. A comparison of the effica-cy of 3 doses of GH (0.025 mg/kg/d,0.05 mg/kg/d, and 0.1 mg/kg/d in pre-pubertal children) found that the 2higher doses resulted in height SDgains of approximately 1.3 at 12months and almost 2 at 24 months. Incontrast, the 0.025 mg/kg/d doseincreased height SD 0.7 and 1.3,respectively.4

Other studies in which the effect ofGH dose on adult height outcome wasevaluated also indicated that a higherdose was more effective. The adultheight of participants of a Genentechcore study using 0.3 mg/kg/wk wascompared to the height outcome fromthe Kabi International Growth Study(KIGS) in which a mean dose of 0.19mg/kg/wk was used. The final heightwas 171.6 cm in the core study versus166.0 cm in KIGS for men and 158.5cm versus 154.9 cm, respectively, forwomen.5,6

The duration of GH therapy alsohas a considerable effect on adultheight outcomes. In 3 studies,5,7,8 theGH dose was constant (0.3 mg/kg/wk)and starting height SDs were similar.The best height outcomes wereobserved in the children treated forthe longest period: 6.4 years in theGenentech clinical trials, 4.6 years inthe National Cooperative GrowthStudy (NCGS), and 3.8 years in theBuffalo study. Mean final heights andgains in height SD for men were 171.6cm (+ 2.3 SD) in the Genentech trials,168 cm (+ 1.3 SD) in NCGS, and 166cm (+1.3 SD) in the Buffalo study.

The frequency of GH administra-tion is another factor that affects finalheight outcome. A 3-year studyinvolving 65 prepubertal children who



were GH deficient on provocative test-ing found that daily administration ofGH resulted in greater height SD gainsat the end of 3 years than did thrice-weekly administration of the samedose (0.3 mg/kg/wk). Over 3 years,there was an extra gain of 1 height SD(or 6.4 cm) with daily treatment.9

Neither bone maturation nor theonset of puberty was accelerated bydaily administration in comparisonwith thrice-weekly administration.

A review of data from theGenentech trials, KIGS, the Buffalostudy, and NCGS revealed that dosesof 0.3 mg/kg/wk have no adverseeffects on bone age or the age of onsetof puberty.10 However, in all 4 studies,the height gain during puberty wasless for females (15 to 19.4 cm) thanfor males, and less for males (22 to 24cm) than for healthy, normally grow-ing children (27 to 30 cm).

A recent study in which a higherdose of GH (0.7 mg/kg/wk) was usedduring puberty reported greater heightgain and growth velocity.11 However,given the cost of GH, it is expectedthat use of the higher dose would berestricted to use in children whobegin treatment late or whose puber-tal growth is suboptimal when stan-dard dosing of GH is administered.

Payer’s Perspective on Diagnosis and Treatment ofPediatric GH Deficiency

Clinicians who submit a requestfor coverage of GH therapy to treatchildren with short stature usuallyhave little idea of how those requestsare processed by payers. In addition,review policies regarding GH therapyvary from payer to payer. The initialpayer response is a review that is usu-ally performed by a nurse or a physi-cian. Some insurance companies havea nurse or physician, who has mini-mal experience with the diagnosis ofGH deficiency in children, reviewingthe request for GH coverage.

In general, insurance companiesprefer concrete policies, and they

should have guidelines for theapproval of GH therapy that havebeen drafted by endocrinologists. Theguidelines should be specific enoughto provide direction but flexibleenough to permit case-by-case deter-minations. Insurance companies alsotend to respond favorably to requeststhat involve uses of GH that areapproved by the Food and DrugAdministration (FDA). However,many insurance companies have theirown policy exclusion criteria that areindependent of FDA approval, whichis their legal right.

Insurance companies also havethe right to gather medical informa-tion, which is essentially the sameinformation that clinicians need tomake their own determination ofmedical necessity. The patient’sgrowth chart and medical history,including laboratory test results suchas scores from provocative tests, arenecessary data. Information on theheight of the patient’s parents shouldbe provided to the payer if that is afactor in the clinician’s decision toprescribe GH.12

In many instances, requests forcoverage of GH therapy are incom-plete, and important data are miss-ing. Often, the growth chart ismissing or is unreadable because offax distortion that “blacks out” thegrid, particularly that portion belowthe 5th percentile and above the95th percentile. Clinicians should beaware of this and be especially care-ful when submitting growth chartsby fax.

Other issues that warrant consid-eration are the minimal and maxi-mal GH dosage, the duration oftherapy, new scientific break-throughs that suggest the benefits ofGH therapy but that have not yetbeen approved by the FDA, and theextent of a new company’s obliga-tion to continue payment for GHtherapy when the insured or theinsured’s parent switches insurancecompanies.



Turner SyndromeTurner syndrome is a common

chromosomal abnormality (45,X0karyotype) that occurs once in every2000 to 5000 live-born phenotypicfemales. The cardinal clinical featuresare short stature and infertility, bothof which are seen in virtually allpatients.

Growth in such individuals isalready subnormal in infancy andearly childhood, and there is nopubertal growth spurt. Becausegrowth velocity declines early on,many affected girls are significantlybelow the 5th percentile for height,even at the age of 2, 3, or 4 years.Because patients with Turner syn-drome are not GH deficient, GH stim-ulation testing is not needed in theUnited States to establish the need forGH therapy as it is in other countries,unless growth is clearly abnormalwhen compared with that expectedfor patients with the syndrome.

In 1995, GH at a pharmacologicdose of 0.375 mg/kg/wk was approvedin the United States for the treatmentof short stature associated withTurner syndrome. With this dose,which is 25% higher than the con-ventional doses used to treat GHdeficiency, many patients canachieve a final height of 60 inches, or152.4 cm.13

Unfortunately, as reflected in theNCGS, early recruitment for thistreatment is lagging; ascertainmentand enrollment peak in patients whorange in age from 10 to 12 years.14

Treatment is thus initiated latebecause it does not allow enough timefor GH therapy to help such patientsachieve a height of 60 inches. In addi-tion, the growth rate for patientsreceiving therapy for Turner syn-drome is not as sustained as with ther-apy for other indications, although thegrowth rates for the first year of treat-ment are comparable. After the firstyear of treatment, the increase ingrowth rate in those with Turner syn-drome begins to decline, which trans-

lates into a less impressive gain inheight SD.

A study by Rosenfeld and associ-ates found that combination therapywith 0.375 mg/kg/wk of GH and 0.625mg/kg/d of oxandrolone resulted in amore robust height gain than did treat-ment with GH alone (152.1 ± 5.9 cm(mean ± SD) versus 150.4 ± 5.5).13

A major issue in the treatment ofTurner syndrome is determiningwhen to introduce estrogen therapy.Clearly, it is not beneficial to intro-duce estrogen too early because it willadvance bone age faster than chrono-logical age and impact negatively onfinal height. Results are better whenGH is given early and estrogen admin-istration is delayed until the patient is15 years of age.15

In summary, the appropriate clini-cal management of Turner syndromeconsists of the early administration ofGH to enable the child to reach amore acceptable adult height earlierand to allow earlier introduction ofestrogen at 12 to 13 years of age if theheight deficit is no longer substantial.Combination therapy with GH and ananabolic agent such as oxandrolonecould be tried if the patient is 10 to 12years old and still has a considerableheight deficit. Baseline measure-ments of follicle-stimulating hormoneand luteinizing hormone providesome clues about whether sponta-neous puberty will occur. If those lev-els are markedly elevated, sponta-neous puberty is unlikely to occur,and anabolic agents can be used withgreater confidence.

Cumulative height changes seen inthe NCGS, in which the duration oftreatment ranged from 1.8 years to5.6 years, clearly demonstrate thatlonger treatment with GH translatesinto greater height gain.13 Recent datafrom a long-term Dutch study inwhich GH doses between 0.46 and0.63 mg/kg/wk were injected daily for7 years show that those high dosesand the longer duration of treatmentresult in final heights of 159.1 cm to




167.2 cm, and that the highest doseused produced the greatest finalheight.16 The height gains observedin the study by Rosenfeld and asso-ciates were not as robust as thoseseen in the Dutch study (probablybecause in the latter trial higherGH doses were used). However,gains in height were more robustthan those seen in the NCGSbecause of the longer duration oftreatment in the study by Rosenfeldand associates (7.6 ± 2.2 years forpatients receiving GH only and 6.1 ±1.9 years for those receiving GH plusoxandrolone).

Insurance coverage for treatmentis likely to be approved easily if onthe statement of medical necessitythe clinician includes the patient’skaryotype and a notation that anopen growth plate has been identi-fied. Some payers explicitly exempttreatment with biologicals, whichinclude GH at low and high doses,from coverage. However, becausemany patients with Turner syndromeare diagnosed late and would there-fore benefit from treatment with high-er doses of GH, payers should allowflexible dosing in individual cases.Because Turner syndrome requiresyears of treatment, payers should alsogrant approval for treatment with GHfor periods of at least 1 year andshould not require that an entire setof clinical data be submitted every 6months.

Chronic Renal InsufficiencyChronic renal insufficiency (CRI)

is characterized by an insidious andirreversible loss of nephron functionand glomerular filtration that pro-gresses to chronic renal failure andend-stage renal disease (ESRD),which necessitates dialysis and/orrenal transplantation. There are noaccurate data on the prevalence ofCRI in the pediatric population, butit is estimated to be much higherthan the 4000 to 4500 cases of ESRDin children.

There are 3 major causes of CRI:

• Structural, which is usually due tomaldevelopment of the urinarytract as a result of a congenitaldefect, as in cases of obstructiveuropathy

• Metabolic, such as cystinosis oroxalosis

• Acquired, which is usually theresult of an infectious process suchas glomerulonephritis, which thenprogresses to glomerulosclerosis

Growth failure is one of the mostcommon clinical features of CRI, andstunted growth is already apparent inmany children at the time of presen-tation. In fact, the younger the age atthe onset of CRI, the greater thedegree of growth failure. Most chil-dren with CRI do not attain theirgenetic height potential (indeed, theirmean adult height is 2 SD below thenorm) because of a combination offactors, including poor growth,delayed puberty that delays andblunts the growth spurt, malnutrition,acidosis, and renal osteodystrophy.Some studies have suggested thatparathyroid hormone levels play themajor role in determining whether achild with CRI will grow normally, butdata are not conclusive.17

The effect of malnutrition ongrowth in children with CRI deservesfurther mention. Often, these chil-dren cannot consume enough caloriesto sustain their catabolic state or grownormally. Although they may grownormally and may maintain a normalgrowth rate when their nutritionalstatus has been optimized and othersupportive measures have been insti-tuted, they do not experience catch-up growth and may not regain theheight they have lost. If those chil-dren are to be treated with GH, theirnutritional and medical status mustbe optimized to ensure an adequateresponse to therapy.

Three mechanisms in the GH-IGFaxis have been implicated in the



poor growth seen in children withCRI:

• A possible decrease in the numberof GH receptors, which reflects adecrease in GH-binding proteinactivity in the serum

• Reduced secretion of IGF-1• An increase in IGFBP-3, which

causes a decrease in the bioavail-ability of free, biologically activeIGF-1

Although it had been thought thatcorrection of the renal status by dial-ysis or transplantation would normal-ize growth in such patients, theresults of studies assessing thesetreatments have been disappointing.In a study of 51 patients undergoingdialysis, one third achieved normalgrowth, one third exhibited retardedgrowth, and one third experiencedseverely retarded growth. Normal,retarded, and severely retardedgrowth are defined as loss of lengthfor age less than .2 SDs/year of thera-py; approximately .39 SDs/year oftherapy; and greater than .5 SDs/yearof therapy, respectively.18 Anotherstudy indicated no improvement ingrowth after continuous ambulatoryperitoneal dialysis.19 Similarly, astudy of children who had undergonerenal transplantation up to 3 yearsearlier found no evidence of catch-upgrowth; their height was maintainedat -2 SD despite normalization ofgrowth rate posttransplant.20

Promising study results of GHtreatment of uremic animals in theearly 1980s prompted successful clin-ical trials in children, which ultimate-ly led to FDA approval of treatmentwith GH before transplantation inthose with CRI. In one study, controlpatients had low-normal IGF-1 levels,and those treated with GH (somat-ropin, in this study) had an improve-ment in IGF-1 levels. At the end of thefirst year of the study, the growth ratewas 6.5 cm/yr in the 28 controlpatients compared with 10.8 cm/yr inthose who received GH. At the end of

the second year, the growth rate was5.5 cm/yr in the control subjects and7.8 cm/yr in the treated patients.When therapy was changed fromplacebo to GH, there was a dramaticimprovement in height SD scoresfrom the pretherapy score of -2.9 SDto –1.9 SD. GH treatment did notaccelerate bone age or the progres-sion to ESRD.21

In summary, treatment with soma-tropin in this study corrects heightdeficit in most children with CRIbefore transplant and significantlyimproves the mean growth rates andheight SD; 65% of treated patientsachieve heights in the normal rangeafter 2 years and 91% achieve normal-range heights after 3 or more years ofsuch therapy.21

Several key issues remain regard-ing CRI in children:

• What are the absolute growth cri-teria for the initiation of GHtherapy pretransplant?

• How soon should GH therapy beinitiated and for how long?

• Should managed care companiesfacilitate optimal medical man-agement before approving GHtreatment?

Idiopathic Short StatureIdiopathic short stature syndrome

is defined as significantly shortstature (greater than -2.5 SD); persist-ently low growth rate; and no evi-dence of systemic disease, malnutri-tion, hypothyroidism, chromosomalabnormality, or classical GH deficien-cy on provocative testing. There aremultiple causes of that syndrome,including genetic short stature, famil-ial short stature, constitutional delayof puberty, a variety of GH axis dys-functions (notably, IGF deficiency),partial GH insensitivity syndrome,combinations of constitutional delayand genetic short stature, and causesof unknown origin.

Dozens of studies reported in theliterature have shown that children



who have been classified as havingidiopathic short stature will have ashort-term increase in growth ratewhen given GH. Long-term studies,most of which were conducted onsmall numbers of patients, have report-ed height gains ranging from none toabout 7.5 cm. Some of the larger stud-ies, however, have shown the mostbenefit from treatment with GH.

In a recently reported study, 121children with idiopathic short staturesyndrome who were prepubertal(mean age, 10 years) at study entrywere treated for 2 to 9 years with GHat a dose of 0.3 mg/kg/wk.22 In anattempt to avoid confounding effectsof puberty, at least for the first fewyears of the study, the investigatorsrecruited girls with a bone age of lessthan 9 years and boys with a bone ageof less than 10 years. Bone age wasdetermined by the Fels method, andpredicted adult height was deter-mined by Bayley-Pinneau tables.

Eighty of the study subjectsachieved more than 98% of their finalheight; the bone age was more than16 years for males and more than 14years for females. The final height was165.5 cm ± 7.2 cm for the 57 malesand 153.1 cm ± 4.8 cm for the 23females, which was halfway betweenpredicted height and midparental tar-get height for those subjects, for a gainover predicted height of 5 cm in malesand 5.9 cm in females. Overall, 50.9%of the males and 60.8% of the femalesreached or exceeded their predictedheight.22

The study findings indicate thatGH treatment increases mean finalheight in those with idiopathic shortstature and that the mean final heightis significantly greater than that ofequally short untreated controls. Twomechanisms may produce the benefi-cial effects:

• GH therapy corrects a defect inGH secretion or action (21% of thestudy subjects had low baselinelevels of GH-binding proteins, and2 subjects had a heterozygous

defect in the extracellular GH-binding region of the GH receptorgene, which suggests partial GHsensitivity; an additional 11% hada low baseline level of IGF-1).

• GH treatment exerts a pharmaco-logic effect on growth.22

The results of the study describedabove and new data from studies con-ducted in England and Germanyindicate that GH treatment has abeneficial effect on adult height inindividuals with idiopathic shortstature. As a group, children with idio-pathic short stature are just as shortas and grow just as slowly as do chil-dren with GH deficiency. The reasonfor this is not fully known. One thirdof the children in the study describedabove had evidence of a disorder ofthe GH-IGF axis.22

GH treatment could be consid-ered for children with idiopathicshort stature who exhibit a signifi-cant height deficiency (-2.5 SD[instead of -2 SD] when comparedwith that of appropriate controls), apersistently low growth rate (lessthan 25% observed over a period ofat least 1 year or less than 40% over2 years), and no evidence of sys-temic disease.

Deciding Coverage for GH Treatment

The need for mutual understandingamong payers and providers regardingauthorization for insurance coveragefor GH treatment has never beenmore urgent. To deal with requestsfrom clinicians to cover GH treat-ment, payers have developed a spec-trum of policy and implementationplans to guide decisions regardingcoverage.

One such policy and implementa-tion plan was developed in 1991 bythe New England-based HarvardPilgrim Health Plan. Under the plan,the GH review committee is chairedby a pediatric endocrinologist andconsists of a medical director, a nursepractitioner/case manager, a pharma-



cy director, and an administrativeassistant/analyst.

When the policy was first imple-mented, the committee approved GHtherapy for the treatment of GH defi-ciency and Turner syndrome. In1996, they approved therapy for CRIprior to transplantation in patientswhose renal failure and nutritionalstatus were managed optimally. Thecommittee added criteria for wastingcaused by the acquired immune defi-ciency syndrome (AIDS) in adults in1997, criteria for GH deficiency inadults in 1998, and pediatric-to-adulttransition criteria (growth velocityless than 2 cm/yr and fused epiphy-seal plates) in 1999.

When committee members devel-oped guidelines for the coverage ofGH treatment, they considered sever-al factors, including safety, efficacy,cost, ethics, and the results ofendocrine studies.

Approval for coverage is usuallygranted in 2 or 3 days for those withGH deficiency or Turner syndromeand for patients with CRI who are toundergo transplantation as long as theapplication meets coverage criteria.

Applications for coverage of GHtherapy in cases in which the criteriafor therapy are not clear requiremore time to process because of con-sultations with experts, surveys, andinitial and secondary reviewing of theliterature.

Another key feature of the policyand implementation plan is therolling review process, which is per-formed every 6 months to determinewhether treatment endpoints arebeing achieved and the therapy isappropriate for the patient.

. . . REFERENCES . . .

1. Boot AM, Engels MA, Boema GJ, et al.Changes in bone mineral density, body com-position, and lipid metabolism during growthhormone (GH) treatment in children withGH deficiency. J Clin Endocrinol Metab1997;82:2423-2428.

2. Kuromaru R, Kohno H, Ueyama N, et al.Long-term prospective study of body compo-sition and lipid profiles during and aftergrowth hormone (GH) treatment in childrenwith GH deficiency: Gender-specific meta-bolic effects. J Clin Endocrinol Metab1998;83:3890-3896.3. Carrel Al, Myers SE, Whitman BY, AllenDB. Growth hormone improves body com-position, fat utilization, physical strength andagility and growth in Prader-Willi syndrome:A controlled study. J Pediatr 1999;134:215-221.4. The American Norditropin Trial Group.Cohen P, Rosenfeld RG, eds. In press.5. Blethen SL, Baptista J, Kuntze J, Foley T,LaFranchi S, Johanson A. Adult height ingrowth hormone (GH)-deficient childrentreated with biosynthetic GH. The GenentechGrowth Study Group. J Clin EndocrinolMetab 1997;82:418-420. 6. Ranke MB, Price DA, Albertsson-WiklandK, Maes M, Lindberg A. Factors determiningpubertal growth and final height in growthhormone treatment of idiopathic growth hor-mone deficiency. Analysis of 195 patients ofthe Kabi Pharmacia International GrowthStudy. Horm Res 1997;48:62-71.7. August GP, Julius JR, Blethen SL. Adultheight in children with growth hormone defi-ciency who are treated with biosyntheticgrowth hormone: The National CooperativeGrowth Study experience. Pediatrics1998;102:512-516.8. MacGillivray MH, Blethen SL, Buchlis JG,Clopper RR, Sandberg DE, Conboy TA.Current dosing of growth hormone in chil-dren with growth hormone deficiency: Howphysiologic? Pediatrics 1998;108:527-530.9. MacGillivray MH, Baptista J, Johanson A.Outcome of a four-year randomized study ofdaily versus three times weekly somatropintreatment in prepubertal naïve growth hor-mone-deficient children. Genentech StudyGroup. J Clin Endocrinol Metab1996;81:1806-1809.10. Sandberg DE, MacGillivray MH. Growthhormone therapy in childhood-onset growthhormone deficiency. Endocrine 2000;12:173-182.11. Mauras N, Reiter EO, Baptista J, Attie KMand the Genentech Study Group. Effects ofhigh rhGH therapy in adolescent childrenwith GH deficiency: A randomized, multi-center study. Pediatr Res 1998;83:81A.12. Katz HP. Growth hormone coverage pol-icy and implementation: A four-year experi-ence. HMO Pract 1997;11:68-73.



13. Rosenfeld RG, Attie KM, Frane J, et al.Growth hormone therapy of Turner’s syn-drome: Beneficial effects on adult height.J Pediatr 1998;132:319-324. 14. Plotnick L, Attie KM, Blethen SL, Sy JP.Growth hormone treatment of girls withTurner syndrome: The National CooperativeGrowth Study experience. Pediatrics1998;102:479-481.15. Chernausek SD, Attie KM, Cara JF,Rosenfeld RG, France J, and the Genentech,Inc. Colloborative Study Group. Growth hor-mone therapy of Turner syndrome: Theimpact of age of estrogen replacement onfinal height. J Clin Endocrinol Metab In press.16. Sas TC, de Muinck Keizer-Schrama SM,Stijnen T, et al. Final height in girls withTurner syndrome treated with once or twicedaily growth hormone injections. DutchAdvisory Group on Growth Hormone. ArchDis Child 1999;80:36-41.17. Fine RN. Growth in children with renalinsufficiency. In: Nissenson A, Fine R,

Gentile D, eds. Clinical Dialysis, 2nd ed.East Norwalk, Conn: Appleton & Lange;1990:676-686.18. Kleinknecht C, Broyer M, GagnadouxMF, et al. Growth in children with long-termdialysis: A study of 76 patients. Adv Nephrol1980;9:133-163.19. von Lilien T, Salusky IB, Bocchat I,Ettenger RB, Fine RN. Five years experiencewith continuous ambulatory and continuouscycling peritoneal dialysis. J Pediatr1997;111:513-518.20. Pennisi AJ, Castin G, Phillips LS, et al.Linear growth in long-term renal allograftrecipients. Clin Nephrol 1977;8:415.21. Nutropin Clinical Trials. Data on file.Genentech, Inc. South San Francisco, CA.22. Hintz RL, Attie KM, Baptista J, Roche A.Effect of growth hormone treatment on adultheight of children with idiopathic shortstature. Genentech Collaborative Group. N Engl J Med 1999;340:502-507.

Documents

Growth Hormone Use in Pediatric Growth Hormone Deficiency ...€¦ · Evaluation for growth hormone (GH) deficiency in a child who is short (ie, whose height is more than 2 standard