Should Survivors of Childhood Cancer or Testicular Cancer Be Screened for Androgen Deficiency?

Alan D. Rogol; Bradley D. Anawalt


Clin Endocrinol. 2018;89(4):397-398. 

Many neoplastic diseases that are common in childhood, adolescence and early adulthood were previously uniformly and swiftly fatal. These neoplastic diseases and their therapies are associated with an increased incidence of multiple endocrinopathies. For example, intracranial neoplasias invade the hypothalamus and/or pituitary and cause deficiencies of growth hormone, gonadotrophins, thyrotrophin or adrenocorticotrophin. Cranial radiation used to treat intracranial neoplasia is often associated with progressive hypopituitarism that may take years to develop into a clinically significant endocrinopathy.[1] Neoplasias that destroy normal endocrine tissue outside of the brain may cause primary endocrinopathies. Testicular cancer, the most common solid malignancy in 18–35 years old men, may cause abnormalities in testicular function by damaging normal testicular histology. Alkylating chemotherapies used in the treatment of testicular cancer commonly cause primary testicular dysfunction.

With the advent of increasingly effective therapies for many of these diseases, many young patients are surviving for decades without recurrent disease.[1,2] Recognition and management of the endocrine sequelae of cancers of childhood and early adulthood are now major considerations in the long-term care of these survivors. In boys and young men, disturbance of normal reproductive function is one of the common, late sequelae of neoplasia or its treatment.[3] The most common long-term manifestation of reproductive dysfunction in male survivors of childhood and early adult neoplastic disease is reduced fertility as an adult. Although most clinicians equate hypogonadism with hypoandrogenaemia, normal gonadal function in men consists of spermatogenesis and production of sex steroid hormones. Hypogonadism may be defined as androgen deficiency, dysspermatogenesis or both.

In conjunction with very high intratesticular testosterone concentrations (100-fold higher than circulating concentrations in men) produced by Leydig cells, Sertoli cells nurture spermatogenesis.[4,5] For normal sperm production, both high intratesticular testosterone concentrations and normal Sertoli cell function are required. It is rare to have significant testosterone deficiency without having some degree of impairment in sperm maturation.

Sertoli cells are more vulnerable to the effects of chemotherapy and testicular and pelvic radiation, and thus, reduced sperm production and fertility are much more common than androgen deficiency in men who are long-term survivors of early-onset cancer.[3] In the Childhood Cancer Survivors Study, the hazard ratio for ever siring a child was 0.56 (95% CI 0.49–0.63) compared with the brothers of the cancer survivors.[6] Sertoli and germ cells for spermatogenesis are more vulnerable after puberty; adolescents and young adults treated for cancer are more likely to be infertile than children treated before puberty.[6]

The prevalence of androgen deficiency in adult survivors of childhood and young adult cancer is unclear. In one study of men who had survived an average of 20 years after the diagnosis and treatment of cancers before age 18, 23% of the survivors had a low serum total testosterone concentration (<10 nmol/L) or an elevated serum LH concentration (>10/IU/L) compared to 4% of controls.[7] However, serum testosterone concentrations were not uniformly measured in the early morning hours. The normal range is based on peak testosterone concentrations that occur in the early morning; serum testosterone concentrations decline significantly after the early morning.[8] In another study of young adult survivors, the prevalence of low early morning serum total testosterone concentrations (<10 nmol/L) was 14% compared to less than 2.5% in controls without cancer.[9]

Most recently, Isaksson and colleagues studied the prevalence of hypogonadism in adult survivors >18 years old) of childhood cancer (diagnosed before age 18) and another cohort of men who had survived testicular cancer diagnosed in early adulthood.[10] These investigators showed a higher prevalence of hypogonadism compared to their control population, all of whom (patients and controls) resided in southern Sweden. They defined a gonadal function outcome based on the following:

  • Primary hypogonadism: total testosterone < 10 nmol/L; LH and FSH > 10 IU/L with FSH > LH concentration; total testosterone < 10 nmol/L, LH ≤ 10 IU/L and FSH > 10 IU/L

  • Secondary hypogonadism: total testosterone <10 nmol/L; LH and FSH ≤ 10 IU/L

  • Compensated hypogonadism: total testosterone ≥ 10 nmol/L but LH > 10 IU/L

  • Current testosterone therapy

For those in the cancer survivor cohort, they found an overall prevalence of hypogonadism of approximately 26% and an odds ratio of 2.1 compared to the control group. They observed that 6% of the cancer survivors had primary hypogonadism, 7% had secondary hypogonadism, 22% had compensated hypogonadism, and another 11% were being treated with testosterone. For the cohort of men who had survived testicular cancer, the prevalence of hypogonadism was 36% with an odds ratio of 6.9 compared to controls. For each subgroup, 14% had primary hypogonadism, 7% had secondary hypogonadism, 7% had compensated hypogonadism, and 10% were being treated with testosterone.

According to the Isaksson study,[10] the prevalence of androgen deficiency is high enough to justify screening in long-term survivors of childhood cancer or testicular cancer in early adulthood. However, the study has several limitations. First, similar to other studies,[7,9] the investigators do not have data on symptoms and signs of androgen deficiency. The diagnosis of androgen deficiency requires symptoms or signs of androgen deficiency.[8] Second, the data in the Isaksson study are based on a single blood sample, and there is enough day-to-day variation of serum testosterone concentrations that it is required that androgen deficiency be confirmed with at least two low serum testosterone concentrations.[8] Third, not all blood samples were obtained in the early morning, and it is unknown whether the men were acutely ill at the time of venipuncture for testosterone measurement. It is well known that measurement of testosterone on a sample after 10 AM or during acute illness may be associated with lower concentrations of testosterone.[8] Fourth, the investigator uses total testosterone concentrations to define androgen deficiency. Many of these men may have had low serum sex hormone-binding globulin concentrations and had normal (calculated) free testosterone concentrations. Men with low total testosterone concentrations and normal calculated free testosterone concentrations are likely eugonadal.[11] Fifth, compensated hypogonadism–a normal serum testosterone concentration, but raised LH concentration–might not be a clinically significant phenomenon. In addition, because LH is secreted in a pulsatile fashion, some men who were categorized as having compensated hypogonadism might have had serum LH concentrations measured during a pulse of LH secretion.[12] Measurement on a second sample may have yielded a normal LH concentration in these men. Finally, the investigators conducted a number of subgroup analyses without evidence of preplanned comparisons. This approach increases the risk of type I and type II statistical errors.

Nonetheless, the study has a number of strengths including relatively large numbers of cancer survivors and a long follow-up period. The data are likely to be complete because the study was based on a Swedish national registry (in a single national healthcare system). The investigators correctly conclude that further studies must be carried out. A prospective, longitudinal cohort study with data on symptoms and signs of androgen deficiency and data on early morning testosterone, FSH and LH concentrations as well as seminal fluid analyses would be invaluable.

In the meantime, it is incumbent on the clinician to enquire about symptoms of testosterone deficiency in men who have survived childhood or testicular cancer and then conduct an appropriate evaluation to determine whether hypogonadism is present.