Abstract

Background

Modern treatment of Hodgkin’s lymphoma (HL) has transformed its prognosis but causes late effects, including premature menopause. Cohort studies of premature menopause risks after treatment have been relatively small, and knowledge about these risks is limited.

Methods

Nonsurgical menopause risk was analyzed in 2127 women treated for HL in England and Wales at ages younger than 36 years from 1960 through 2004 and followed to 2003 through 2012. Risks were estimated using Cox regression, modified Poisson regression, and competing risks. All statistical tests were two-sided.

Results

During follow-up, 605 patients underwent nonsurgical menopause before age 40 years. Risk of premature menopause increased more than 20-fold after ovarian radiotherapy, alkylating chemotherapy other than dacarbazine, or BEAM (bis-chloroethylnitrosourea [BCNU], etoposide, cytarabine, melphalan) chemotherapy for stem cell transplantation, but was not statistically significantly raised after adriamycin, bleomycin, vinblastine, dacarbazine (ABVD). Menopause generally occurred sooner after ovarian radiotherapy (62.5% within five years of ≥5 Gy treatment) and BEAM (50.9% within five years) than after alkylating chemotherapy (24.2% within five years of ≥6 cycles), and after treatment at older than at younger ages. Cumulative risk of menopause by age 40 years was 81.3% after greater than or equal to 5Gy ovarian radiotherapy, 75.3% after BEAM, 49.1% after greater than or equal to 6 cycles alkylating chemotherapy, 1.4% after ABVD, and 3.0% after solely supradiaphragmatic radiotherapy. Tables of individualized risk information for patients by future period, treatment type, dose and age are provided.

Conclusions

Patients treated with HL need to plan intended pregnancies using personalized information on their risk of menopause by different future time points.

Modern treatments have transformed the prognosis of Hodgkin’s lymphoma (HL), but at the cost of risk of several long-term ill effects, including gonadal failure. In women this can take the form of acute temporary ovarian failure or of premature menopause. The latter, as well as resulting in loss of reproductive function and ovarian sex hormone secretion, can adversely affect risks of cardiovascular disease ( 1 , 2 ), osteoporosis (3. 4 ), and possibly ischaemic stroke ( 5 ), although it can also reduce the risk of radiotherapy-induced breast cancer ( 6 ).

Type of treatment, notably alkylating chemotherapy ( 7–9 ) and pelvic radiotherapy ( 7 , 8 , 10 ), has been found to influence the risk of premature menopause after HL. It has been unclear whether age at treatment also affects risk ( 7–9 ). For personalized advice to patients, however, so that they can plan their future pregnancies, risk data are needed individualized to several factors—current age, the specific duration ahead of interest, the age at which treatment occurred, and type of treatment. No such analysis has been published, reflecting the relatively small size of studies to date—the largest that included detailed treatment data were based on 65 ( 8 ), 97 ( 9 ), and 104 ( 7 ) patients with nonsurgical premature menopause. We therefore analyzed data on women treated for HL under 36 years of age in England and Wales from 1960 through 2004, of whom 605 underwent premature, nonsurgical menopause during follow-up.

Methods

The study was based on a national clinical recall exercise in England and Wales, started in November 2003, for women in the country who had been treated with supradiaphragmatic radiotherapy for HL under 36 years of age. Several sources of patient identification were used to ensure that all patients in the country treated since 1971, or an earlier year depending on the region, were identified. Complete histories of treatment for HL were obtained from case notes. For all subjects who were still alive and had not emigrated at the time of the clinical recall exercise, questionnaires were handed out at clinics or sent by mail, unless the consultant felt that this was clinically inappropriate for a particular patient. The questionnaire, with signed informed consent, asked about various environmental, behavioral, menstrual, and reproductive factors, including age at menopause and cause of menopause, and potential confounders such as smoking and anthropometric variables. Further details about case identification, ethical approval, and data collection can be found in Swerdlow et al. ( 11 ).

Statistical Analysis

Cohort analyses were undertaken to analyze risk of premature menopause in relation to treatment type, age at treatment, and other predictor variables. Details of person-years calculations are given in the Supplementary Materials (available online). Women whose menopause occurred before they were treated for HL or who received at least 30 Gy radiation to the brain (and hence may have developed menopause through radiation-induced gonadotrophin deficiency) were excluded from the analysis.

We analyzed separately the effects of each common chemotherapy regimen, but also aggregated, for larger numbers, all alkylating regimens other than ABVD, and have termed these “classic alkylating chemotherapy.” We excluded ABVD because previous research ( 9 ) and our own data suggest that its effect on the ovary is far less than that of other common alkylating regimens.

Ovarian doses from infradiaphragmatic radiotherapy were estimated by simulation ( Supplementary Materials , available online).

To estimate the relative risk of menopause for women at ages under 40 in relation to various treatment variables, we calculated hazard ratios by Cox regression ( 12 ). These analyses were adjusted for age and for treatments other than that under analysis; the effect of further adjustment for potential confounding by smoking, body mass index (BMI), and parity ( 13 ) was also examined.

Risk ratios for premature menopause in relation to age at treatment were estimated by modified Poisson regression with robust standard errors ( 14 ). Cumulative risks of reaching menopause by various ages were estimated by the competing risks method ( 15 ). Trends in hazard ratios were assessed by a likelihood ratio test, trends in risk ratios by a Wald test, and trends in cumulative incidence by a Wald test applied to an arcsine root transformation. All variables were analyzed as categorical, using the cut points shown in the tables. All P values cited are two-sided; P less than .05 was deemed to be statistically significant. Analyses were conducted using Stata ( 16 ). Further details are in Supplementary Methods (available online).

Results

Five thousand and two women were identified who had received supradiaphragmatic radiotherapy for HL in England and Wales from 1960 through 2004. Of these, 1105 had died, and 35 had emigrated before the clinical recall process; for 47 no current address could be identified, for 53 the consultant deemed it inappropriate for a questionnaire to be sent, and for 255 the consultant could not be identified or mailing was not possible for other reasons. The remaining 3507 were invited to complete a questionnaire, of whom 2509 (72%) returned a completed questionnaire, 280 replied that they were willing but never actually returned the questionnaire, 32 declined, and 686 did not reply. We excluded from analysis six respondents because their menopause occurred before HL treatment or as a consequence of ovarian transposition at the time of treatment, or they had received at least 30 Gy brain radiation, 128 because they did not state their menopausal status (52) or had reached menopause at an unstated age (76), and 248 because they were on oral contraceptives or hormone replacement therapy (HRT) throughout follow-up. This left 2127 subjects in the analysis. Five and a half percent of these subjects were younger than age 15 years at first treatment, 50.0% aged 15 to 24 years, and 44.5% aged 25 to 35 years ( Table 1 ). The subjects were first treated from 1960 through 2004, mainly in the 1980s (33.6%) and 1990s (36.4%). The questionnaires were completed from 2003 through 2012, 0.3 to 48.4 (median = 17.8) years after first treatment.

Table 1.

Descriptive characteristics of study subjects

Characteristic No. (%) 
Age at first treatment, y 
 0–14 117 (5.5) 
 15–19 441 (20.7) 
 20–24 622 (29.2) 
 25–29 518 (24.3) 
 30–35 429 (20.2) 
Year of first treatment 
 1960–69 66 (3.1) 
 1970–79 413 (19.4) 
 1980–89 714 (33.6) 
 1990–99 775 (36.4) 
 2000–04 159 (7.5) 
Year of questionnaire completion 
 2003–05 843 (39.6) 
 2006–08 1148 (54.0) 
 2009–12 136 (6.4) 
Duration from first treatment
to questionnaire completion, y  
 0–9 444 (20.9) 
 10–19 791 (37.2) 
 20–29 594 (27.9) 
 30–39 274 (12.9) 
 ≥40 24 (1.1) 
Treatment 
No pelvic RT, no alkylating CT* 624 (29.3) 
Pelvic RT, no alkylating CT 55 (2.6) 
Classic alkylating CT†, no pelvic RT‡ 931 (43.8) 
ABVD, no other alkylating CT, no pelvic RT 144 (6.8) 
BEAM, no pelvic RT 58 (2.7) 
Transplant with other or unknown regimen, no pelvic RT 46 (2.2) 
Pelvic RT and classic alkylating CT or ABVD or BEAM or other transplant regimen‡ 119 (5.6) 
Unknown§ 150 (7.1) 
Total 2127 (100) 
Characteristic No. (%) 
Age at first treatment, y 
 0–14 117 (5.5) 
 15–19 441 (20.7) 
 20–24 622 (29.2) 
 25–29 518 (24.3) 
 30–35 429 (20.2) 
Year of first treatment 
 1960–69 66 (3.1) 
 1970–79 413 (19.4) 
 1980–89 714 (33.6) 
 1990–99 775 (36.4) 
 2000–04 159 (7.5) 
Year of questionnaire completion 
 2003–05 843 (39.6) 
 2006–08 1148 (54.0) 
 2009–12 136 (6.4) 
Duration from first treatment
to questionnaire completion, y  
 0–9 444 (20.9) 
 10–19 791 (37.2) 
 20–29 594 (27.9) 
 30–39 274 (12.9) 
 ≥40 24 (1.1) 
Treatment 
No pelvic RT, no alkylating CT* 624 (29.3) 
Pelvic RT, no alkylating CT 55 (2.6) 
Classic alkylating CT†, no pelvic RT‡ 931 (43.8) 
ABVD, no other alkylating CT, no pelvic RT 144 (6.8) 
BEAM, no pelvic RT 58 (2.7) 
Transplant with other or unknown regimen, no pelvic RT 46 (2.2) 
Pelvic RT and classic alkylating CT or ABVD or BEAM or other transplant regimen‡ 119 (5.6) 
Unknown§ 150 (7.1) 
Total 2127 (100) 

* ie, solely supradiaphragmatic radiotherapy ± non alkylating CT. ABVD = adriamycin, bleomycin, vinblastine, dacarbazine; BCNU = bis-chloroethylnitrosourea; BEAM = BCNU, etoposide, cytarabine, melphalan; CT = chemotherapy; RT = radiotherapy.

† Classic alkylating agents received in the study were: mustine, melphalan, chlorambucil, procarbazine, cyclophosphamide, BCNU, lomustine, and ifosfamide. Alkylating agents overall included these plus dacarbazine.

‡ Includes some patients treated with pelvic RT and classic alkylating CT who did not meet the criteria for analysis in Table 2 .

§ ie, supradiaphragmatic radiotherapy, but not known if any other treatment in addition.

During 25936 person-years of follow-up (mean 12.2 years per woman), 880 patients reached menopause (including 71 surgical menopause), 70 were censored at breast cancer diagnosis, 967 remained premenopausal, and 210 were on contraceptives or HRT at the end of follow-up, with consequent uncertainty about menopause. Seven subjects, treated between ages 11 and 15 years, never underwent menarche; we have categorized these for analytical purposes with menopause due to treatment, because all had been followed to at least age 19 years, and the consequences for their future reproduction were the same as menopause. In 687 subjects, nonsurgical menopause was early (ie, age <45 years) and in 605 premature (ie, age <40 years). Compared with women who had received neither alkylating chemotherapy nor pelvic radiotherapy, risk of premature nonsurgical menopause was increased about 20-fold in women who had received pelvic radiotherapy or classic alkylating chemotherapy, 36-fold in women who had received both, and over 50-fold in those who had received a stem cell transplant (more again if they had also received pelvic radiotherapy). Risk was not statistically significantly raised, however, after ABVD ( Table 2 ).

Table 2.

Risks of menopause under age 40 years in relation to modality and dose of treatment

Modality and dose of treatment No. of patients No. with menopause <age 40y Hazard ratio* (95% CI) P 
Treatment 
 No alkylating CT, no pelvic RT 624 17 1.0  
 Pelvic RT, no alkylating CT 55 23 23.0 (12.3 to 43.1) <.001 
 Classic alkylating CT, no pelvic RT† 931 364 20.2 (12.4 to 32.9) <.001 
 Classic alkylating CT and pelvic RT† 68 40 35.6 (20.1 to 62.9) <.001 
 ABVD, no pelvic RT‡ 144 0.7 (0.2 to 2.9) .59 
 ABVD and pelvic RT‡ 30.4 (4.0 to 230.6) <.001 
 BEAM, no pelvic RT 58 38 60.6 (34.1 to 107.8) <.001 
 BEAM and pelvic RT 125.9 (45. 7 to 346.7) <.001 
 Other or unknown transplant regimen, no pelvic RT 46 32 53.8 (29.7 to 97.2) <.001 
 Other or unknown transplant regimen and pelvic RT 75.4 (22.0 to 258.4) <.001 
 Unknown 150 50 15.4 (8.9 to 26.7) <.001 
Ovarian RT dose§ 
 0 Gy|| 1871 469 1.0  
 <5 Gy 92 34 1.5 (1.1 to 2.1) .03 
 5–29 Gy 6.4 (3.0 to 13.6) <.001 
 ≥30 Gy 24 23 14.1 (9.2 to 21.7) <.001 
 Unknown 49 17 2.0 (1.2 to 3.3) .01 
Ptrend   <.001  
Cycles of classic alkylating CT¶     
 0 cycles 758 19 1.0  
 1–3 cycles 79 21 13.0 (7.0 to 24.1) <.001 
 4–5 cycles 111 36 16.0 (9.2 to 27.9) <.001 
 6 cycles 362 153 24.1 (15.0 to 38.9) <.001 
 7–12 cycles 210 81 21.3 (12.9 to 35.2) <.001 
 >12 cycles 10 187.8 (77.5 to 454.8) <.001 
 Unknown no. cycles 146 55 21.5 (12.8 to 36.3) <.001 
Ptrend   <.001  
Cycles of ABVD¶     
 0 cycles 624 17 1.0  
 1–5 cycles 38 2.7 (0.3 to 21.3) .35 
 6 cycles 71 1.4 (0.2 to 11.7) .73 
 ≥7 cycles 25 0.0 (0.0 to 0.0) 
 Unknown no. cycles 10 0.0 (0.0 to 0.0) 
Ptrend   .99  
Chemotherapy regimen 
 No alkylating CT, no pelvic RT 624 17 1.0  
 ABVD¶ 144 1.3 (0.3 to 6.2) .73 
 ChlVPP¶ 383 150 22.7 (13.8 to 37.4) <.001 
 LOPP¶ 273 95 17.9 (10.8 to 29.9) <.001 
 MVPP¶ 115 66 37.1 (21.2 to 65.0) <.001 
 MOPP¶ 91 38 19.3 (10.5 to 35.3) <.001 
 BEAM +any other classic alkylating CT, no pelvic RT 53 37 97.9 (52.9 to 181.0) <.001 
 BEAM but no other classic alkylating CT, no pelvic RT 8.4 (1.0 to 73.9) .06 
Modality and dose of treatment No. of patients No. with menopause <age 40y Hazard ratio* (95% CI) P 
Treatment 
 No alkylating CT, no pelvic RT 624 17 1.0  
 Pelvic RT, no alkylating CT 55 23 23.0 (12.3 to 43.1) <.001 
 Classic alkylating CT, no pelvic RT† 931 364 20.2 (12.4 to 32.9) <.001 
 Classic alkylating CT and pelvic RT† 68 40 35.6 (20.1 to 62.9) <.001 
 ABVD, no pelvic RT‡ 144 0.7 (0.2 to 2.9) .59 
 ABVD and pelvic RT‡ 30.4 (4.0 to 230.6) <.001 
 BEAM, no pelvic RT 58 38 60.6 (34.1 to 107.8) <.001 
 BEAM and pelvic RT 125.9 (45. 7 to 346.7) <.001 
 Other or unknown transplant regimen, no pelvic RT 46 32 53.8 (29.7 to 97.2) <.001 
 Other or unknown transplant regimen and pelvic RT 75.4 (22.0 to 258.4) <.001 
 Unknown 150 50 15.4 (8.9 to 26.7) <.001 
Ovarian RT dose§ 
 0 Gy|| 1871 469 1.0  
 <5 Gy 92 34 1.5 (1.1 to 2.1) .03 
 5–29 Gy 6.4 (3.0 to 13.6) <.001 
 ≥30 Gy 24 23 14.1 (9.2 to 21.7) <.001 
 Unknown 49 17 2.0 (1.2 to 3.3) .01 
Ptrend   <.001  
Cycles of classic alkylating CT¶     
 0 cycles 758 19 1.0  
 1–3 cycles 79 21 13.0 (7.0 to 24.1) <.001 
 4–5 cycles 111 36 16.0 (9.2 to 27.9) <.001 
 6 cycles 362 153 24.1 (15.0 to 38.9) <.001 
 7–12 cycles 210 81 21.3 (12.9 to 35.2) <.001 
 >12 cycles 10 187.8 (77.5 to 454.8) <.001 
 Unknown no. cycles 146 55 21.5 (12.8 to 36.3) <.001 
Ptrend   <.001  
Cycles of ABVD¶     
 0 cycles 624 17 1.0  
 1–5 cycles 38 2.7 (0.3 to 21.3) .35 
 6 cycles 71 1.4 (0.2 to 11.7) .73 
 ≥7 cycles 25 0.0 (0.0 to 0.0) 
 Unknown no. cycles 10 0.0 (0.0 to 0.0) 
Ptrend   .99  
Chemotherapy regimen 
 No alkylating CT, no pelvic RT 624 17 1.0  
 ABVD¶ 144 1.3 (0.3 to 6.2) .73 
 ChlVPP¶ 383 150 22.7 (13.8 to 37.4) <.001 
 LOPP¶ 273 95 17.9 (10.8 to 29.9) <.001 
 MVPP¶ 115 66 37.1 (21.2 to 65.0) <.001 
 MOPP¶ 91 38 19.3 (10.5 to 35.3) <.001 
 BEAM +any other classic alkylating CT, no pelvic RT 53 37 97.9 (52.9 to 181.0) <.001 
 BEAM but no other classic alkylating CT, no pelvic RT 8.4 (1.0 to 73.9) .06 

* All adjusted for age at treatment; ovarian RT dose adjusted for classic alkylating CT; classic alkylating CT cycles and ABVD cycles adjusted for pelvic RT. Hazard ratios estimated by Cox regression. All statistical tests were two-sided. ABVD = adriamycin, bleomycin, vinblastine, dacarbazine; BCNU = bis-chloroethylnitrosourea; BEAM = BCNU, etoposide, cytarabine, melphalan; ChlVPP = chlorambucil, vinblastine, procarbazine,prednisone; CI = confidence interval; CT = chemotherapy; LOPP = chlorambucil, vincristine, procarbazine, prednisone; MOPP = mechlorethamine, vincristine, procarbazine, prednisone; MVPP = mechlorethamine, vinblastine, procarbazine, prednisone; RT = radiotherapy.

† No BEAM, no other transplant regimen.

‡ No classic alkylating CT, no BEAM, no other transplant regimen.

§ Including individuals who received alkylating CT, or for whom it is unknown whether they received alkylating CT, as well as RT, and adjusted for alkylating CT. This approach was taken because there were too few patients who received pelvic RT without alkylating CT, to analyze dose response in this category.

|| Only supradiaphragmatic radiotherapy.

¶ Category is individuals who received this regimen and no other alkylating regimen and no pelvic RT.

Risk of premature menopause increased with number of cycles of classic alkylating chemotherapy (P < .001) and with ovarian dose from radiotherapy (P < .001), but not with number of cycles of ABVD. The results in Table 2 are shown for risk of menopause younger than age 40 years, but the results were similar if the outcome considered was menopause younger than age 35 years or younger than age 45 years (data not shown). They were also not materially changed by adjustment for smoking, BMI, and parity or by changes in the form of the model (combining or removing categories with small numbers).

Risk was greatly raised after treatment with each common classic alkylating regimen ( Table 2 ), greatest for mechlorethamine, vinblastine, procarbazine, prednisone (hazard ratio [HR] = 37.1, 95% CI = 21.2 to 65.0, P < .001), and least for chlorambucil, vincristine, procarbazine, prednisone (HR = 17.9, 95% CI = 10.8 to 29.9, P < .001). Risk was greater again after BEAM, the most common regimen for stem cell transplantation; almost all BEAM-treated patients had also received other classic alkylating treatments. Risk was also statistically significantly raised in patients treated with etoposide who had not received classic alkylating chemotherapy or pelvic radiotherapy (HR = 6.1, 95% CI = 1.7 to 21.4, P < .001).

Risks increased moderately but statistically significantly ( P < .001) with increasing age at treatment ( Table 3 ). A similar effect was present separately for patients who received classic alkylating chemotherapy without pelvic radiotherapy ( P < .001), but was not apparent for other categories of patient, based on smaller numbers. Risks were not related to temporal proximity of treatment to menarche (data not shown).

Table 3.

Risks of menopause under age 40 years by age at treatment and type of treatment

Treatment*, age at first treatment, y  No. of
patients†  
No. with menopause <age 40y Risk ratio (95% CI) P 
≥6 cycles classic alkylating CT, no pelvic RT, no BEAM, no other transplant regimen‡ 
  0–19 45 15 1.00  
  20–24 84 46 1.63 (1.03 to 2.57) .04 
  25–29 97 65 1.99 (1.29 to 3.09) .002 
  30–35 102 68 1.98 (1.28 to 3.06) .002 
Ptrend   <.001  
All treatments‡ 
  0–19 189 54 1.00  
  20–24 319 119 1.29 (1.01 to 1.64) .04 
  25–29 320 142 1.44 (1.14 to 1.81) .002 
  30–35 333 142 1.45 (1.15 to 1.82) .002 
Ptrend   .001  
Treatment*, age at first treatment, y  No. of
patients†  
No. with menopause <age 40y Risk ratio (95% CI) P 
≥6 cycles classic alkylating CT, no pelvic RT, no BEAM, no other transplant regimen‡ 
  0–19 45 15 1.00  
  20–24 84 46 1.63 (1.03 to 2.57) .04 
  25–29 97 65 1.99 (1.29 to 3.09) .002 
  30–35 102 68 1.98 (1.28 to 3.06) .002 
Ptrend   <.001  
All treatments‡ 
  0–19 189 54 1.00  
  20–24 319 119 1.29 (1.01 to 1.64) .04 
  25–29 320 142 1.44 (1.14 to 1.81) .002 
  30–35 333 142 1.45 (1.15 to 1.82) .002 
Ptrend   .001  

* Specific treatment categories other than ≥ 6 cycles classic alkylating CT not shown because of small numbers. Risk ratios estimated by modified Poisson regression with robust standard errors. All statistical tests were two-sided. BCNU = bis-chloroethylnitrosourea; BEAM = BCNU, etoposide, cytarabine, melphalan; CI = confidence interval; CT = chemotherapy; RT = radiotherapy.

† Numbers are much smaller than in Tables 1 , 2 , 4 , and 5 because calculation of risk ratios can only include subjects who completed the questionnaire at age ≥40 years and had a known menopausal status (ie, not on oral contraceptives or hormone replacement therapy and no hysterectomy) at age 40 years.

‡ Adjusted for ovarian radiotherapy dose, cycles of classic alkylating chemotherapy, or both, as appropriate.

Table 4 shows cumulative risks of menopause up to various attained ages in relation to type of treatment and age at treatment. Cumulative risk of menopause by age 40 years was 81.3% after at least 5Gy ovarian radiotherapy, 75.3% after BEAM, 49.1% after at least 6 cycles alkylating chemotherapy, 1.4% after ABVD, and 3.0% after solely supradiaphragmatic radiotherapy. Ovarian radiotherapy of 5Gy or greater at any age tended to lead rapidly to menopause, especially if combined with classic alkylating chemotherapy: for instance, 62.5% of those treated with at least 5 Gy and 78.6% of those treated with at least 5 Gy radiotherapy plus at least 6 cycles of classic alkylating chemotherapy were menopausal by age 35 years, compared with only 1.2% of those treated solely with supradiaphragmatic radiotherapy. After classic alkylating chemotherapy without pelvic radiotherapy, the effect was more gradual: 29.2% were menopausal by age 35 years but this increased to 49.1% by age 40 years and 63.0% by age 45 years. BEAM had rapid effects on menopause, and ABVD appeared to give effects late if at all. Analyses by duration since treatment (data not shown) also showed more rapid effects of ovarian radiotherapy (62.5% menopausal five years after treatment with at least 5 Gy, 87.5% at 20 years) and BEAM (50.9% at five years, 70.5% at 20 years) than classic alkylating chemotherapy (24.2% at five years after at least 6 cycles, 51.4% at 20 years), and more rapid effects after older than younger-age treatment (eg, 7.8% five years after treatment at ages 0 to 19 years; 33.1% five years after treatment at ages 30 to 35 years).

Table 4.

Cumulative risk of menopause up to different attained ages, by type of treatment and age at treatment*

Type of treatment† Age at treatment, y Cumulative risk (95% CI) at attained age, y 
25 30 35 40 45 
No alkylating CT, no pelvic RT 0–19 191 0.0 0.0 0.7 (0.1 to 3.5) 3.1 (1.0 to 7.2) 5.2 (2.1 to 10.5) 
 20–24 177 0.0 0.6 (0.1 to 2.9) 1.2 (0.2 to 3.9) 4.1 (1.7 to 8.2) 5.8 (2.7 to 10.6) 
 25–29 146  1.4 (0.3 to 4.4) 2.1 (0.6 to 5.5) 3.0 (1.0 to 7.1) 6.2 (2.7 to 11.9) 
 30–35 110   0.9 (0.1 to 4.5) 1.9 (0.4 to 6.0) 5.4 (2.0 to 11.5) 
 All ages 624   1.2 (0.5 to 2.3) 3.0 (1.8 to 4.8) 5.6 (3.7 to 8.0) 
Ptrend    .63 .28 .68 
≥5Gy ovarian RT, no alkylating CT 0–19 100.0 100.0 100.0 100.0 100.0 
 20–24 20.0 (0.8 to 58.2) 40.0 (5.2 to 75.3) 40.0 (5.2 to 75.3) 80.0 (20.4 to 96.9) 80.0 (20.4 to 96.9) 
 25–29  75.0 (12.8 to 96.1) 75.0 (12.8 to 96.1) 75.0 (12.8 to 96.1) 75.0 (12.8 to 96.1) 
 30–35   60.0 (12.6 to 88.2) 80.0 (20.4 to 96.9) 100.0 
 All ages 16   62.5 (34.9 to 81.1) 81.3 (52.5 to 93.5) 87.5 (58.6 to 96.7) 
Ptrend    .63 1.00 .38 
≥6 cycles classic alkylating CT, no pelvic RT 0–19 159 9.5 (5.6 to 14.7) 10.8 (6.6 to 16.3) 12.5 (7.8 to 18.4) 24.9 (17.2 to 33.4) 42.9 (30.0 to 55.1) 
 20–24 165 10.3 (6.3 to 15.5) 17.1 (11.8 to 23.2) 25.5 (18.9 to 32.5) 45.9 (36.6 to 54.7) 55.4 (44.5 to 65.1) 
 25–29 153  17.9 (12.3 to24.4) 36.0 (28.4 to 43.7) 58.1 (48.8 to 66.2) 72.2 (60.5 to 81.0) 
 30–35 120   42.5 (33.6 to 51.1) 65.3 (55.9 to 73.2) 80.3 (70.0 to 87.3) 
 All ages 597   29.2 (25.5 to 33.1) 49.1 (44.4 to 53.5) 63.0 (57.4 to 68.0) 
Ptrend    .02 .03 .01 
≥5 Gy ovarian RT, ≥6 cycles classic alkylating CT 0–19 100.0 100.0 100.0 100.0 100.0 
 20–24 71.4 (25.8 to 92.0) 85.7 (33.4 to 97.9) 85.7 (33.4 to 97.9) 85.7 (33.4 to 97.9) 85.7 (33.4 to 97.9) 
 25–29  75.0 (12.8 96.1) 75.0 (12.8 to 96.1) 100.0 100.0 
 30–35   50.0 (0.6 to 91.0) 100.0 100.0 
 All ages 14   78.6 (47.2 to 92.5) 92.9 (59.1 to 99.0) 92.9 (59.1 to 99.0) 
Ptrend    
≥6 cycles ABVD, no other alkylating CT, no pelvic RT 0–19 13 0.0 0.0 0.0 0.0 0.0 
 20–24 26 0.0 0.0 7.0 (0.4 to 27.1) 7.0 (0.4 to 27.1) 7.0 (0.4 to 27.1) 
 25–29 24  0.0 0.0 0.0 0.0 
 30–35 33   0.0 0.0 27.0 (3.2 to 60.9) 
 All ages 96   1.4 (0.1 to 6.8) 1.4 (0.1 to 6.8) 20.1 (3.2 to 47.3) 
Ptrend    
BEAM, no pelvic RT 0–19 14 30.0 (9.2 to 54.5) 45.7 (18.9 to 69.3) 45.7 (18.9 to 69.3) 61.4 (25.8 to 83.9) 61.4 (25.8 to 83.9) 
 20–24 17 29.4 (10.7 to 51.1) 58.8 (32.5to 77.8) 58.8 (32.5 to 77.8) 94.1 (65.0 to 99.1) 94.1 (65.0 to 99.1) 
 25–29 17  37.0 (15.1 to 59.2) 85.8 (54.3 to 96.2) 85.8 (54.3 to 96.2) 85.8 (54.3 to 96.2) 
 30–35 10   50.0 (18.4 to 75.3) 70.0 (32.9 to 89.2) 70.0 (32.9 to 89.2) 
 All ages 58   61.7 (47.3 to 73.2) 75.3 (58.3 to 86.2) 75.3 (58.3 to 86.2) 
Ptrend    .64 .91 .91 
All treatments 0–19 556 8.8 (6.6 to11.3) 10.6 (8.2 to 13.4) 13.1 (10.4 to 16.2) 19.9 (16.2 to 23.9) 27.6 (22.7 to 32.7) 
 20–24 621 8.7 (6.7 to 11.1) 14.8 (12.2 to 17.8) 19.9 (16.8 to 23.3) 31.4 (27.3 to 35.5) 38.2 (33.5 to42.9) 
 25–29 515  14.6 (11.7 to 17.8) 25.1 (21.4 to 29.0) 37.9 (33.4 to 42.4) 47.2 (41.9 to 52.3) 
 30–35 429   24.2 (20.2 to 28.3) 40.3 (35.5 to 45.0) 50.5 (45.1 to 55.7) 
 All ages 2121   21.0 (19.2 to 22.8) 32.8 (30.6 to 35.0) 41.2 (38.7 to 43.8) 
Ptrend    .09 .05 .03 
Type of treatment† Age at treatment, y Cumulative risk (95% CI) at attained age, y 
25 30 35 40 45 
No alkylating CT, no pelvic RT 0–19 191 0.0 0.0 0.7 (0.1 to 3.5) 3.1 (1.0 to 7.2) 5.2 (2.1 to 10.5) 
 20–24 177 0.0 0.6 (0.1 to 2.9) 1.2 (0.2 to 3.9) 4.1 (1.7 to 8.2) 5.8 (2.7 to 10.6) 
 25–29 146  1.4 (0.3 to 4.4) 2.1 (0.6 to 5.5) 3.0 (1.0 to 7.1) 6.2 (2.7 to 11.9) 
 30–35 110   0.9 (0.1 to 4.5) 1.9 (0.4 to 6.0) 5.4 (2.0 to 11.5) 
 All ages 624   1.2 (0.5 to 2.3) 3.0 (1.8 to 4.8) 5.6 (3.7 to 8.0) 
Ptrend    .63 .28 .68 
≥5Gy ovarian RT, no alkylating CT 0–19 100.0 100.0 100.0 100.0 100.0 
 20–24 20.0 (0.8 to 58.2) 40.0 (5.2 to 75.3) 40.0 (5.2 to 75.3) 80.0 (20.4 to 96.9) 80.0 (20.4 to 96.9) 
 25–29  75.0 (12.8 to 96.1) 75.0 (12.8 to 96.1) 75.0 (12.8 to 96.1) 75.0 (12.8 to 96.1) 
 30–35   60.0 (12.6 to 88.2) 80.0 (20.4 to 96.9) 100.0 
 All ages 16   62.5 (34.9 to 81.1) 81.3 (52.5 to 93.5) 87.5 (58.6 to 96.7) 
Ptrend    .63 1.00 .38 
≥6 cycles classic alkylating CT, no pelvic RT 0–19 159 9.5 (5.6 to 14.7) 10.8 (6.6 to 16.3) 12.5 (7.8 to 18.4) 24.9 (17.2 to 33.4) 42.9 (30.0 to 55.1) 
 20–24 165 10.3 (6.3 to 15.5) 17.1 (11.8 to 23.2) 25.5 (18.9 to 32.5) 45.9 (36.6 to 54.7) 55.4 (44.5 to 65.1) 
 25–29 153  17.9 (12.3 to24.4) 36.0 (28.4 to 43.7) 58.1 (48.8 to 66.2) 72.2 (60.5 to 81.0) 
 30–35 120   42.5 (33.6 to 51.1) 65.3 (55.9 to 73.2) 80.3 (70.0 to 87.3) 
 All ages 597   29.2 (25.5 to 33.1) 49.1 (44.4 to 53.5) 63.0 (57.4 to 68.0) 
Ptrend    .02 .03 .01 
≥5 Gy ovarian RT, ≥6 cycles classic alkylating CT 0–19 100.0 100.0 100.0 100.0 100.0 
 20–24 71.4 (25.8 to 92.0) 85.7 (33.4 to 97.9) 85.7 (33.4 to 97.9) 85.7 (33.4 to 97.9) 85.7 (33.4 to 97.9) 
 25–29  75.0 (12.8 96.1) 75.0 (12.8 to 96.1) 100.0 100.0 
 30–35   50.0 (0.6 to 91.0) 100.0 100.0 
 All ages 14   78.6 (47.2 to 92.5) 92.9 (59.1 to 99.0) 92.9 (59.1 to 99.0) 
Ptrend    
≥6 cycles ABVD, no other alkylating CT, no pelvic RT 0–19 13 0.0 0.0 0.0 0.0 0.0 
 20–24 26 0.0 0.0 7.0 (0.4 to 27.1) 7.0 (0.4 to 27.1) 7.0 (0.4 to 27.1) 
 25–29 24  0.0 0.0 0.0 0.0 
 30–35 33   0.0 0.0 27.0 (3.2 to 60.9) 
 All ages 96   1.4 (0.1 to 6.8) 1.4 (0.1 to 6.8) 20.1 (3.2 to 47.3) 
Ptrend    
BEAM, no pelvic RT 0–19 14 30.0 (9.2 to 54.5) 45.7 (18.9 to 69.3) 45.7 (18.9 to 69.3) 61.4 (25.8 to 83.9) 61.4 (25.8 to 83.9) 
 20–24 17 29.4 (10.7 to 51.1) 58.8 (32.5to 77.8) 58.8 (32.5 to 77.8) 94.1 (65.0 to 99.1) 94.1 (65.0 to 99.1) 
 25–29 17  37.0 (15.1 to 59.2) 85.8 (54.3 to 96.2) 85.8 (54.3 to 96.2) 85.8 (54.3 to 96.2) 
 30–35 10   50.0 (18.4 to 75.3) 70.0 (32.9 to 89.2) 70.0 (32.9 to 89.2) 
 All ages 58   61.7 (47.3 to 73.2) 75.3 (58.3 to 86.2) 75.3 (58.3 to 86.2) 
Ptrend    .64 .91 .91 
All treatments 0–19 556 8.8 (6.6 to11.3) 10.6 (8.2 to 13.4) 13.1 (10.4 to 16.2) 19.9 (16.2 to 23.9) 27.6 (22.7 to 32.7) 
 20–24 621 8.7 (6.7 to 11.1) 14.8 (12.2 to 17.8) 19.9 (16.8 to 23.3) 31.4 (27.3 to 35.5) 38.2 (33.5 to42.9) 
 25–29 515  14.6 (11.7 to 17.8) 25.1 (21.4 to 29.0) 37.9 (33.4 to 42.4) 47.2 (41.9 to 52.3) 
 30–35 429   24.2 (20.2 to 28.3) 40.3 (35.5 to 45.0) 50.5 (45.1 to 55.7) 
 All ages 2121   21.0 (19.2 to 22.8) 32.8 (30.6 to 35.0) 41.2 (38.7 to 43.8) 
Ptrend    .09 .05 .03 

* The table shows the future risk of menopause during follow-up to various attained ages, from the viewpoint of a patient at the time of treatment. Cumulative risks (%) estimated by the competing risks method. All statistical tests were two-sided. ABVD = adriamycin, bleomycin, vinblastine, dacarbazine; BCNU = bis-chloroethylnitrosourea; BEAM = BCNU, etoposide, cytarabine, melphalan; CI = confidence interval; CT = chemotherapy; RT = radiotherapy.

† The particular treatment groups analyzed are selected as those homogeneous groups of greatest interest, so do not total to the “all treatments” category.

‡ No BEAM, no other transplant regimen.

Table 5 and Figure 1 show estimated future cumulative risks of menopause over various future periods for a patient currently of a particular age and who has not yet reached menopause, according to treatment history. For instance, for a patient who is now aged 30 years and not yet menopausal and was treated with ≥6 cycles of classic alkylating chemotherapy between ages 0 and 19 years, the risk of menopause in the next 10 years is 20.0%, whereas for a similar patient treated at ages 25 to 29 years it is 49.4%, and for a patient of the same age who was treated with BEAM it is 50.0% or more, depending on their age at treatment. We have presented risks for periods of 10 to 25 years ahead rather than shorter periods because fertility is lost in the general population about 10 years before menopause occurs ( 17 ), so if the same is true in women treated for HL, risks of menopause less than 10 years ahead are immaterial to risk of future loss of fertility.

Table 5.

Estimated cumulative risk of menopause over various future time periods in relation to current age, type of treatment, and age at treatment*

Type of treatment† Age at treatment, y Current age 20 years  
Cumulative risk (95% CI) at future risk period 
10 years 15 years 20 years 25 years 
No alkylating CT, no pelvic RT 0–19 169 0.0   0.8 (0.1 to 3.9) 3.5 (1.1 to 8.1) 5.9 (2.4 to 11.7)  
20–24               
25–29               
30–35               
All ages 169 0.0   0.8 (0.1 to 3.9) 3.5 (1.1 to 8.1) 5.9 (2.4 to 11.7)  
≥5 Gy ovarian RT, no alkylating CT All ages 100.0   100.0   100.0   100.0    
≥6 cycles classic alkylating CT, no pelvic RT 0–19 144 7.1 (3.6 to 12.1) 9.0 (4.9 to 14.6) 22.7 (14.7 to 31.7) 42.5 (28.5 to 55.7)  
20–24               
25–29               
30–35               
All ages 144 7.1 (3.6 to 12.1) 9.0 (4.9 to 14.6) 22.7 (14.7 to 31.7) 42.5 (28.5 to 55.7)  
≥5 Gy ovarian RT, ≥6 cycles classic alkylating CT ‡  All ages 100.0   100.0   100.0   100.0    
≥6 cycles ABVD, no pelvic RT §  0–19 13 0.0   0.0   0.0   0.0    
20–24               
25–29               
30–35               
All ages 13 0.0   0.0   0.0   0.0    
BEAM, no pelvic RT 0–19 12 45.0 (16.4 to 70.3) 45.0 (16.4 to 70.3) 63.3 (23.1 to 86.7) 63.3 (23.1 to 86.7)  
20–24               
25–29               
30–35               
All ages 12 45.0 (16.4 to 70.3) 45.0 (16.4 to 70.3) 63.3 (23.1 to 86.7) 63.3 (23.1 to 86.7)  
All treatments 0–19 486 6.5 (4.5 to 9.0) 9.4 (6.8 to 12.3) 17.1 (13.3 to 21.3) 25.8 (20.5 to 31.4)  
20–24               
25–29               
30–35               
All ages 486 6.5 (4.5 to 9.0) 9.4 (6.8 to 12.3) 17.1 (13.3 to 21.3) 25.8 (20.5 to 31.4)  
Type of treatment †  Age at treatment, y  Current age 25 years  
 Cumulative risk (95% CI) at future risk period 
n 10 years 15 years 20 years 
No alkylating CT, no pelvic RT 0–19 161 0.8 (0.1 to 4.1) 3.6 (1.2 to 8.3) 6.1 (2.4 to 12.1)     
20–24 174 1.2 (0.2 to 4.0) 4.2 (1.7 to 8.3) 5.9 (2.7 to 10.8)     
25–29               
30–35               
All ages 335 1.0 (0.3 to 2.7) 3.9 (2.0 to 6.8) 5.9 (3.4 to 9.5)     
≥5 Gy ovarian RT, no alkylating CT All ages 25.0 (0.9 to 66.5) 75.0 (12.8 to 96.1) 75.0 (12.8 to 96.1)     
≥6 cycles classic alkylating CT, no pelvic RT 0–19 120 3.8 (1.2 to 8.9) 19.7 (11.4 to 29.5) 42.5 (26.8 to 57.3)     
20–24 147 16.9 (11.1 to 23.7) 39.6 (30.0 to 49.1) 50.3 (38.6 to 60.9)     
25–29               
30–35               
All ages 267 11.0 (7.5 to 15.4) 30.9 (24.2 to 37.9) 46.2 (36.8 to 55.0)     
≥5 Gy ovarian RT, ≥6 cycles classic alkylating CT ‡  All ages 33.3 (0.9 to 77.4) 66.7 (5.4 to 94.5) 66.7 (5.4 to 94.5)     
≥6 cycles ABVD, no pelvic RT §  0–19 0.0   0.0   0.0       
20–24 24 7.3 (0.5 to 28.1) 7.3 (0.5 to 28.1) 7.3 (0.5 to 28.1)     
25–29               
30–35               
All ages 28 7.3 (0.5 to 28.1) 7.3 (0.5 to 28.1) 7.3 (0.5 to 28.1)     
BEAM, no pelvic RT 0–19 33.3 (4.6 to 67.6) 66.7 (7.7 to 93.8) 66.7 (7.7 to 93.8)     
20–24 11 45.5 (16.7 to 70.7) 100.0   100.0       
25–29               
30–35               
All ages 17 41.2 (18.6 to 62.6) 77.9 (23.5 to 95.8) 77.9 (23.5 to 95.8)     
All treatments 0–19 411 5.5 (3.5 to 8.2) 14.1 (10.4 to 18.5) 23.8 (18.2 to 29.9)     
20–24 557 12.4 (9.7 to 15.4) 25.4 (21.3 to 29.7) 33.0 (28.0 to 37.9)     
25–29               
30–35               
All ages 968 9.5 (7.7 to 11.6) 20.7 (17.8 to 23.8) 29.1 (25.4 to 32.9)     
Type of treatment †  Age at treatment, y n Current age 30 years n Current age 35 years 
Cumulative risk (95% CI) at future risk period Cumulative risk (95% CI) at future risk period 
10 years 15 years 10 years 15 years 
No alkylating CT, no pelvic RT 0–19 144 3.7 (1.2 to 8.5) 6.2 (2.5 to 12.4) 113 5.6 (2.1 to 11.9) 17.2 (8.3 to 28.7) 
20–24 162 3.7 (1.4 to 7.8) 5.5 (2.4 to 10.4) 143 4.9 (2.0 to 9.8) 18.7 (10.3 to 29.0) 
25–29 144 1.7 (0.3 to 5.4) 4.9 (1.8 to 10.5) 129 4.4 (1.4 to 10.0) 18.5 (10.0 to 29.1) 
30–35        107 4.6 (1.5 to 10.6) 17.7 (9.2 to 28.4) 
All ages 450 3.0 (1.6 to 5.1) 5.5 (3.4 to 8.4) 492 4.9 (3.0 to 7.4) 18.1 (13.5 to 23.3) 
≥5 Gy ovarian RT, no alkylating CT All ages 50.0 (5.8 to 84.5) 50.0 (5.8 to 84.5) 66.7 (19.5 to 90.4) 66.7 (19.5 to 90.4) 
≥6 cycles classic alkylating CT, no pelvic RT ‡  0–19 103 20.0 (11.2 to 30.7) 45.4 (28.0 to 61.3) 75 46.8 (28.0 to 63.6) 65.8 (42.4 to 81.5) 
20–24 129 35.5 (25.5 to 45.6) 47.3 (34.9 to 58.8) 90 41.9 (28.4 to 54.8) 73.6 (45.6 to 88.7) 
25–29 124 49.4 (39.1 to 58.9) 66.8 (53.0 to 77.3) 79 59.3 (41.8 to 73.1) 88.5 (71.5 to 95.6) 
30–35        68 66.7 (50.7 to 78.5) 78.6 (61.2 to 88.9) 
All ages 356 36.5 (30.5 to 42.5) 53.5 (45.4 to 60.9) 312 53.5 (45.5 to 60.8) 76.8 (63.6 to 85.7) 
≥5 Gy ovarian RT, ≥6 cycles classic alkylating CT ‡  All ages 60.0 (12.6 to 88.2) 60.0 (12.6 to 88.2) 75.0 (12.8 to 96.1) 75.0 (12.8 to 96.1) 
≥6 cycles ABVD, no pelvic RT §  0–19 
20–24 16 7.7 (0.5 to 29.2) 7.7 (0.5 to 29.2) 0.0 (0.0)    
25–29 20 0.0 0.0 10 0.0   0.0 
30–35        30 27.0 (3.2 to 60.9) 54.4 (9.4 to 85.5) 
All ages 36 3.0 (0.2 to 13.4) 3.0 (0.2 to 13.4) 44 20.1 (2.7 to 49.3) 42.1 (7.2 to 75.2) 
BEAM, no pelvic RT 0–19 50.0 (0.6 to 91.0) 50.0 (0.6 to 91.0) 50.0 (0.6 to 91.0) 50.0 (0.6 to 91.0) 
20–24 100.0 100.0 100.0 100.0 
25–29 10 77.5 (35.7 to 93.9) 77.5 (35.7 to 93.9) 0.0 0.0 
30–35        50.0 (5.8 to 84.5) 50.0 (5.8 to 84.5) 
All ages 20 66.9 (27.8 to 88.1) 66.9 (27.8 to 88.1) 10 50.5 (12.9 to 79.6) 50.5 (12.9 to 79.6) 
All treatments 0–19 350 12.7 (8.9 to 17.1) 23.1 (17.2. to 29.4) 257 21.1 (15.0 to 27.8) 34.6 (25.8 to 43.6) 
20–24 476 20.4 (16.4 to 24.8) 28.8 (23.7 to 34.0) 360 24.4 (19.3 to 29.9) 41.6 (33.5 to 49.4) 
25–29 431 27.5 (22.9 to 32.2) 38.4 (32.7 to 44.1) 313 30.2 (24.2 to 36.4) 43.1 (35.1 to 50.8) 
30–35        309 35.7 (29.6 to 41.9) 47.7 (39.9 to 55.1) 
All ages 1257 20.8 (18.3 to 23.5) 30.6 (27.4 to 33.9) 1239 28.2 (25.2 to 31.2) 42.1 (38.0 to 46.1) 
Type of treatment† Age at treatment, y Current age 20 years  
Cumulative risk (95% CI) at future risk period 
10 years 15 years 20 years 25 years 
No alkylating CT, no pelvic RT 0–19 169 0.0   0.8 (0.1 to 3.9) 3.5 (1.1 to 8.1) 5.9 (2.4 to 11.7)  
20–24               
25–29               
30–35               
All ages 169 0.0   0.8 (0.1 to 3.9) 3.5 (1.1 to 8.1) 5.9 (2.4 to 11.7)  
≥5 Gy ovarian RT, no alkylating CT All ages 100.0   100.0   100.0   100.0    
≥6 cycles classic alkylating CT, no pelvic RT 0–19 144 7.1 (3.6 to 12.1) 9.0 (4.9 to 14.6) 22.7 (14.7 to 31.7) 42.5 (28.5 to 55.7)  
20–24               
25–29               
30–35               
All ages 144 7.1 (3.6 to 12.1) 9.0 (4.9 to 14.6) 22.7 (14.7 to 31.7) 42.5 (28.5 to 55.7)  
≥5 Gy ovarian RT, ≥6 cycles classic alkylating CT ‡  All ages 100.0   100.0   100.0   100.0    
≥6 cycles ABVD, no pelvic RT §  0–19 13 0.0   0.0   0.0   0.0    
20–24               
25–29               
30–35               
All ages 13 0.0   0.0   0.0   0.0    
BEAM, no pelvic RT 0–19 12 45.0 (16.4 to 70.3) 45.0 (16.4 to 70.3) 63.3 (23.1 to 86.7) 63.3 (23.1 to 86.7)  
20–24               
25–29               
30–35               
All ages 12 45.0 (16.4 to 70.3) 45.0 (16.4 to 70.3) 63.3 (23.1 to 86.7) 63.3 (23.1 to 86.7)  
All treatments 0–19 486 6.5 (4.5 to 9.0) 9.4 (6.8 to 12.3) 17.1 (13.3 to 21.3) 25.8 (20.5 to 31.4)  
20–24               
25–29               
30–35               
All ages 486 6.5 (4.5 to 9.0) 9.4 (6.8 to 12.3) 17.1 (13.3 to 21.3) 25.8 (20.5 to 31.4)  
Type of treatment †  Age at treatment, y  Current age 25 years  
 Cumulative risk (95% CI) at future risk period 
n 10 years 15 years 20 years 
No alkylating CT, no pelvic RT 0–19 161 0.8 (0.1 to 4.1) 3.6 (1.2 to 8.3) 6.1 (2.4 to 12.1)     
20–24 174 1.2 (0.2 to 4.0) 4.2 (1.7 to 8.3) 5.9 (2.7 to 10.8)     
25–29               
30–35               
All ages 335 1.0 (0.3 to 2.7) 3.9 (2.0 to 6.8) 5.9 (3.4 to 9.5)     
≥5 Gy ovarian RT, no alkylating CT All ages 25.0 (0.9 to 66.5) 75.0 (12.8 to 96.1) 75.0 (12.8 to 96.1)     
≥6 cycles classic alkylating CT, no pelvic RT 0–19 120 3.8 (1.2 to 8.9) 19.7 (11.4 to 29.5) 42.5 (26.8 to 57.3)     
20–24 147 16.9 (11.1 to 23.7) 39.6 (30.0 to 49.1) 50.3 (38.6 to 60.9)     
25–29               
30–35               
All ages 267 11.0 (7.5 to 15.4) 30.9 (24.2 to 37.9) 46.2 (36.8 to 55.0)     
≥5 Gy ovarian RT, ≥6 cycles classic alkylating CT ‡  All ages 33.3 (0.9 to 77.4) 66.7 (5.4 to 94.5) 66.7 (5.4 to 94.5)     
≥6 cycles ABVD, no pelvic RT §  0–19 0.0   0.0   0.0       
20–24 24 7.3 (0.5 to 28.1) 7.3 (0.5 to 28.1) 7.3 (0.5 to 28.1)     
25–29               
30–35               
All ages 28 7.3 (0.5 to 28.1) 7.3 (0.5 to 28.1) 7.3 (0.5 to 28.1)     
BEAM, no pelvic RT 0–19 33.3 (4.6 to 67.6) 66.7 (7.7 to 93.8) 66.7 (7.7 to 93.8)     
20–24 11 45.5 (16.7 to 70.7) 100.0   100.0       
25–29               
30–35               
All ages 17 41.2 (18.6 to 62.6) 77.9 (23.5 to 95.8) 77.9 (23.5 to 95.8)     
All treatments 0–19 411 5.5 (3.5 to 8.2) 14.1 (10.4 to 18.5) 23.8 (18.2 to 29.9)     
20–24 557 12.4 (9.7 to 15.4) 25.4 (21.3 to 29.7) 33.0 (28.0 to 37.9)     
25–29               
30–35               
All ages 968 9.5 (7.7 to 11.6) 20.7 (17.8 to 23.8) 29.1 (25.4 to 32.9)     
Type of treatment †  Age at treatment, y n Current age 30 years n Current age 35 years 
Cumulative risk (95% CI) at future risk period Cumulative risk (95% CI) at future risk period 
10 years 15 years 10 years 15 years 
No alkylating CT, no pelvic RT 0–19 144 3.7 (1.2 to 8.5) 6.2 (2.5 to 12.4) 113 5.6 (2.1 to 11.9) 17.2 (8.3 to 28.7) 
20–24 162 3.7 (1.4 to 7.8) 5.5 (2.4 to 10.4) 143 4.9 (2.0 to 9.8) 18.7 (10.3 to 29.0) 
25–29 144 1.7 (0.3 to 5.4) 4.9 (1.8 to 10.5) 129 4.4 (1.4 to 10.0) 18.5 (10.0 to 29.1) 
30–35        107 4.6 (1.5 to 10.6) 17.7 (9.2 to 28.4) 
All ages 450 3.0 (1.6 to 5.1) 5.5 (3.4 to 8.4) 492 4.9 (3.0 to 7.4) 18.1 (13.5 to 23.3) 
≥5 Gy ovarian RT, no alkylating CT All ages 50.0 (5.8 to 84.5) 50.0 (5.8 to 84.5) 66.7 (19.5 to 90.4) 66.7 (19.5 to 90.4) 
≥6 cycles classic alkylating CT, no pelvic RT ‡  0–19 103 20.0 (11.2 to 30.7) 45.4 (28.0 to 61.3) 75 46.8 (28.0 to 63.6) 65.8 (42.4 to 81.5) 
20–24 129 35.5 (25.5 to 45.6) 47.3 (34.9 to 58.8) 90 41.9 (28.4 to 54.8) 73.6 (45.6 to 88.7) 
25–29 124 49.4 (39.1 to 58.9) 66.8 (53.0 to 77.3) 79 59.3 (41.8 to 73.1) 88.5 (71.5 to 95.6) 
30–35        68 66.7 (50.7 to 78.5) 78.6 (61.2 to 88.9) 
All ages 356 36.5 (30.5 to 42.5) 53.5 (45.4 to 60.9) 312 53.5 (45.5 to 60.8) 76.8 (63.6 to 85.7) 
≥5 Gy ovarian RT, ≥6 cycles classic alkylating CT ‡  All ages 60.0 (12.6 to 88.2) 60.0 (12.6 to 88.2) 75.0 (12.8 to 96.1) 75.0 (12.8 to 96.1) 
≥6 cycles ABVD, no pelvic RT §  0–19 
20–24 16 7.7 (0.5 to 29.2) 7.7 (0.5 to 29.2) 0.0 (0.0)    
25–29 20 0.0 0.0 10 0.0   0.0 
30–35        30 27.0 (3.2 to 60.9) 54.4 (9.4 to 85.5) 
All ages 36 3.0 (0.2 to 13.4) 3.0 (0.2 to 13.4) 44 20.1 (2.7 to 49.3) 42.1 (7.2 to 75.2) 
BEAM, no pelvic RT 0–19 50.0 (0.6 to 91.0) 50.0 (0.6 to 91.0) 50.0 (0.6 to 91.0) 50.0 (0.6 to 91.0) 
20–24 100.0 100.0 100.0 100.0 
25–29 10 77.5 (35.7 to 93.9) 77.5 (35.7 to 93.9) 0.0 0.0 
30–35        50.0 (5.8 to 84.5) 50.0 (5.8 to 84.5) 
All ages 20 66.9 (27.8 to 88.1) 66.9 (27.8 to 88.1) 10 50.5 (12.9 to 79.6) 50.5 (12.9 to 79.6) 
All treatments 0–19 350 12.7 (8.9 to 17.1) 23.1 (17.2. to 29.4) 257 21.1 (15.0 to 27.8) 34.6 (25.8 to 43.6) 
20–24 476 20.4 (16.4 to 24.8) 28.8 (23.7 to 34.0) 360 24.4 (19.3 to 29.9) 41.6 (33.5 to 49.4) 
25–29 431 27.5 (22.9 to 32.2) 38.4 (32.7 to 44.1) 313 30.2 (24.2 to 36.4) 43.1 (35.1 to 50.8) 
30–35        309 35.7 (29.6 to 41.9) 47.7 (39.9 to 55.1) 
All ages 1257 20.8 (18.3 to 23.5) 30.6 (27.4 to 33.9) 1239 28.2 (25.2 to 31.2) 42.1 (38.0 to 46.1) 

* A graphical illustration of data from this table, to facilitate its clinical use, is given in Figure 1. ABVD = adriamycin, bleomycin, vinblastine, dacarbazine; BCNU = bis-chloroethylnitrosourea; BEAM = BCNU, etoposide, cytarabine, melphalan; CI = confidence interval; CT = chemotherapy; RT = radiotherapy.

† The particular treatment groups analyzed are selected as those homogeneous groups of greatest interest, so do not total to the “All treatments” category. Cumulative risks (%) estimated by the competing risks method. All statistical tests were two-sided.

‡ No BEAM, no other transplant regimen.

§ No classic alkylating CT, no BEAM, no other transplant regimen.

Figure 1.

Illustration of use of Table 5 to give patient information on future cumulative risk of menopause: future risk for patients currently aged 25 who were treated at age 20, for three treatment types. Cumulative risks estimated by the competing risks method. All statistical tests were two-sided. BEAM = BCNU (bis-chloroethylnitrosourea), etoposide, cytarabine, melphalan.

Figure 1.

Illustration of use of Table 5 to give patient information on future cumulative risk of menopause: future risk for patients currently aged 25 who were treated at age 20, for three treatment types. Cumulative risks estimated by the competing risks method. All statistical tests were two-sided. BEAM = BCNU (bis-chloroethylnitrosourea), etoposide, cytarabine, melphalan.

Discussion

Premature menopause is an important long-term side effect of treatment for HL, but one on which there has been relatively little systematic information. There has been only one sizeable study in HL patients, 518 patients in the Netherlands, 97 of whom developed premature menopause during follow-up ( 9 ), but this was restricted to chemotherapy patients. Two studies of childhood cancer patients included a minority of cases with HL, with some analyses for this subgroup ( 7 , 8 ). Furthermore, with the exception of one study of children ( 8 ), the studies have largely or entirely included surgical as well as nonsurgical menopause, ie, have potentially diluted apparent risks by including menopause that could not be due to treatment as well as menopause that might be treatment-related.

We found very high risks of premature nonsurgical menopause after each of classic alkylating chemotherapy and pelvic radiotherapy, and greater again after both combined, as seen previously ( 7–10 ). Our radiation dose analyses were based on a relatively crude assessment of ovarian dose compared with those by Sklar et al. ( 8 ), but very similar to that used for most patients by de Bruin et al. ( 9 ). This probably made little difference, however, because our analyses, like theirs, considered dose in a small number of categories, so that imprecision of dose assessment would only matter if it resulted in erroneous allocation by category. However, only 6% of estimated doses were within 10% of a boundary between categories, so modest errors would have made little difference. In accordance with Sklar et al. ( 8 ) and de Bruin et al. ( 9 ), we found increasing risk of premature menopause with increasing dose of classic alkylating chemotherapy. We also found a steep gradient of risk with radiotherapy dose to the ovaries, which was not seen, but based on very small numbers, in the only previous study to compare different doses ( 8 ).

There have been very limited previous data comparing risks of menopause from different chemotherapeutic regimens ( 9 , 18 , 19 ). We found greatly raised risks from each alkylating regimen except ABVD, and particularly large risks after BEAM and other transplant regimens, although these patients had received other classic alkylating regimens in addition. The only previous risk analysis for ABVD was based on 25 treated patients and did not find raised risk ( 9 ); small descriptive studies have varied in the proportion of ABVD-treated patients who reported amenorrhoea ( 20 ). Risk of premature menopause after etoposide without alkylating treatments does not appear to have been investigated previously, but risk was statistically significantly raised in our data, and transient ovarian failure after etoposide has been reported in several patients ( 21 ). The risk would therefore be worth reexamination elsewhere.

Although three large studies have compared the effects of pelvic radiotherapy and alkylating chemotherapy for childhood cancer overall ( 7 , 8 , 10 ), none have examined this for HL, and there is no information on the comparative time courses of menopause occurrence after these treatments. We found that although each can give rise to large risks of eventual premature menopause, the time courses are very different—in general menopause occurred rapidly after radiotherapy and after BEAM, but much more slowly after chemotherapy. This has important practical implications for patients: individual decisions on reproductive planning need to be based on the likely time course of occurrence of menopause, rather than the risk by age 40 years or some other arbitrary age.

The effect of age at treatment on risk of premature menopause has been unclear (whereas for acute ovarian failure, a greater effect of older treatment has been shown [ 19 ]). Childhood cancer studies ( 7 , 8 ) have given variable results, based on small numbers, and the only analysis in HL patients ( 9 ) stated that cumulative risks were “not much different” by age, but based on small numbers and not tested statistically. Our data, based on much larger numbers, show a highly statistically significant but moderate trend of greater risk with older age at treatment. Such a trend might be explicable from the decline in oocyte numbers with age. Radiotherapy and alkylating chemotherapy deplete antral follicles ( 22 , 23 ), and if they destroy a particular percentage of oocytes this will leave fewer oocytes (and hence may more often trigger menopause, which occurs when about 1000 oocytes remain [24]) after therapy at an older than at a younger age ( 25 ). The effect of age in our data was present after classic alkylating chemotherapy, but not apparent based on small numbers after radiotherapy or BEAM.

Risks of menopause within a few years after treatment, unlike eventual risk by age 40 years, were much greater if treatment was at older than at younger ages, reflecting the longer period over which premature menopause could occur in a patient treated at young ages than at older.

Even after supradiaphragmatic radiotherapy, the ovarian dose from which is very low ( 9 ), the risk of premature menopause in our HL patients (4% overall) was probably above that in the general population (1% has been found in a US study [ 26 ]). Previously, Sklar et al. ( 8 ) found a nine-fold greater risk of premature menopause after HL than other childhood cancers after controlling for treatment. The raised risk in our patients might indicate an effect of HL per se, but also might, at least partly, reflect scatter doses of radiation during supradiaphragmatic radiotherapy, which has been estimated at a mean of 0.2 Gy based on individual simulations ( 9 ), or that some nonalkylating chemotherapies may have an effect, as suggested by our findings for etoposide.

Although our study was far larger than any previously conducted and included detailed treatment histories and excluded surgical menopause, it also had limitations. It was based on treatments prevalent several years ago; this is inevitable in any study of this question, because long-term follow-up over many years is by definition needed before risks of menopause many years after treatment can be assessed. Because it was based on a cohort of patients treated with supradiaphragmatic radiotherapy, we could not directly assess risks of menopause in that minority of women who received chemotherapy or pelvic radiotherapy without supradiaphragmatic radiotherapy. For comparisons of risk ratios between pelvic radiotherapy or chemotherapy subgroups, this should not have made any material difference, because all of the subgroups would have received supradiaphragmatic radiotherapy. Our estimates of cumulative risks after pelvic radiotherapy or chemotherapy, however, might be expected to be slightly greater than in patients who do not receive supradiaphragmatic radiotherapy, although not greatly so because the effect of supradiaphragmatic radiotherapy appears to have been relatively slight if any.

Our study was based on a questionnaire several years after treatment, and hence inevitably omitted patients who died early or who did not respond to the questionnaire. This is the same as in other major studies ( 7 , 8 ), except the study by de Bruin et al.( 9 ), which was based on case-note data. The similarity of our results to those of de Bruin ( 9 ), for the subset of analyses also undertaken by de Bruin, suggests that the questionnaire approach was not seriously biased. Also, it seems unlikely that nonresponse in our study was biased by menopausal status in relation to treatment, especially as the questionnaire was sent to patients as an investigation of breast cancer risk, not of menopause per se.

The risk of infertility is an important issue that needs discussion with patients of reproductive age with neoplasms such as Hodgkin’s lymphoma ( 27–29 ). The data provided here should help to provide quantitative underpinning for such consultations and for decisions on fertility preservation.

In conclusion, we have provided for the first time tables to give individualized information for women treated for HL at young ages on their risk of menopause by different ages and over different future risk periods according to their type, dose, and age at treatment. The data show the importance of giving information on the likely future time course of menopause risk, rather than solely the risk by a single future arbitrary age such as 40 years.

Funding

The work was supported by Breakthrough Breast Cancer (BR 02/04); and the European Commission (223497). The Institute of Cancer Research and Royal Marsden Hospital acknowledge funding from the National Institute for Health Research to the Biomedical Research Centre.

The study sponsors had no role in the design of the study, the collection, analysis, or interpretation of the data, the writing of the manuscript, nor the decision to submit the manuscript for publication.

We thank the health staff, cancer networks, cancer registries, and breast screening centers who contributed data, the patients, the Institute of Cancer Research staff who worked on the study, and Justine Windsor, Cheryl Cavanagh, Michael Richards, Ausra Kesminiene, Danielle Morris, Michael Jones, and Alan Ashworth for advice.

References

1.
Kannel
WB
Hjortland
MC
McNamara
PM
et al
Menopause and risk of cardiovascular disease: the Framingham study
.
Ann Intern Med
  .
1976
;
85
(
4
):
447
452
.
2.
de Kleijn
MJ
van der Schouw
YT
Verbeek
AL
et al
Endogenous estrogen exposure and cardiovascular mortality risk in postmenopausal women
.
Am J Epidemiol
  .
2002
;
155
(
4
):
339
345
.
3.
Lindquist
O
Bengtsson
C
Hansson
T
et al
Age at menopause and its relation to osteoporosis
.
Maturitas
  .
1979
;
1
(
3
):
175
181
.
4.
Vega
EM
Egea
MA
Mautalen
CA
.
Influence of the menopausal age on the severity of osteoporosis in women with vertebral fractures
.
Maturitas
  .
1994
;
19
(
2
):
117
124
.
5.
Lisabeth
LD
Beiser
AS
Brown
DL
et al
Age at natural menopause and risk of ischemic stroke: the Framingham heart study
.
Stroke
  .
2009
;
40
(
4
):
1044
1049
.
6.
van Leeuwen
FE
Klokman
WJ
Stovall
M
et al
Roles of radiation dose, chemotherapy, and hormonal factors in breast cancer following Hodgkin’s disease
.
J Natl Cancer Inst
  .
2003
;
95
(
13
):
971
980
.
7.
Byrne
J
Fears
TR
Gail
MH
et al
Early menopause in long-term survivors of cancer during adolescence
.
Am J Obstet Gynecol
  .
1992
;
166
(
3
):
788
793
.
8.
Sklar
CA
Mertens
AC
Mitby
P
et al
Premature menopause in survivors of childhood cancer: a report from the childhood cancer survivor study
.
J Natl Cancer Inst
  .
2006
;
98
(
13
):
890
896
.
9.
De Bruin
ML
Huisbrink
J
Hauptmann
M
et al
Treatment-related risk factors for premature menopause following Hodgkin lymphoma
.
Blood
  .
2008
;
111
(
1
):
101
108
.
10.
Chiarelli
AM
Marrett
LD
Darlington
G
.
Early menopause and infertility in females after treatment for childhood cancer diagnosed in 1964–1988 in Ontario, Canada
.
Am J Epidemiol
  .
1999
;
150
(
3
):
245
254
.
11.
Swerdlow
AJ
Cooke
R
Bates
A
et al
Breast Cancer Risk After Supradiaphragmatic Radiotherapy for Hodgkin’s Lymphoma in England and Wales: A National Cohort Study
.
J Clin Oncol
  .
2012
;
30
(
22
):
2745
2752
.
12.
Cox
DR
.
Regression models and life-tables (with Discussion)
.
J R Stat Soc
  .
1972
;
B34
:
187
220
.
13.
Morris
DH
Jones
ME
Schoemaker
MJ
et al
Body mass index, exercise, and other lifestyle factors in relation to age at natural menopause: analyses from the Breakthrough Generations Study
.
Am J Epidemiol
  .
2012
;
175
(
10
):
998
1005
.
14.
Zou
G
.
A modified poisson regression approach to prospective studies with binary data
.
Am J Epidemiol
  .
2004
;
159
(
7
):
702
706
.
15.
Kalbfleisch
JD
Prentice
RL
.
The Statistical Analysis of Failure Time Data
  .
New York, NY
:
John Wiley & Sons
;
1980
.
16.
StataCorp
.
Stata Statistical Software: Release 10. 2007
  .
College Station, TX
,
Stata Corporation
.
17.
Broekmans
FJ
Soules
MR
Fauser
BC
.
Ovarian aging: mechanisms and clinical consequences
.
Endocr Rev
  .
2009
;
30
(
5
):
465
493
.
18.
Clark
ST
Radford
JA
Crowther
D
et al
Gonadal function following chemotherapy for Hodgkin’s disease: a comparative study of MVPP and a seven-drug hybrid regimen
.
J Clin Oncol
  .
1995
;
13
(
1
):
134
139
.
19.
Behringer
K
Breuer
K
Reineke
T
et al
Secondary amenorrhea after Hodgkin’s lymphoma is influenced by age at treatment, stage of disease, chemotherapy regimen, and the use of oral contraceptives during therapy: a report from the German Hodgkin’s Lymphoma Study Group
.
J Clin Oncol
  .
2005
;
23
(
30
):
7555
7564
.
20.
Hodgson
DC
Pintilie
M
Gitterman
L
et al
Fertility among female Hodgkin lymphoma survivors attempting pregnancy following ABVD chemotherapy
.
Hematol Oncol
  .
2007
;
25
(
1
):
11
15
.
21.
Choo
YC
Chan
SY
Wong
LC
et al
Ovarian dysfunction in patients with gestational trophoblastic neoplasia treated with short intensive courses of etoposide (VP-16–213)
.
Cancer
  .
1985
;
55
(
10
):
2348
2352
.
22.
Larsen
EC
Muller
J
Schmiegelow
K
et al
Reduced ovarian function in long-term survivors of radiation- and chemotherapy-treated childhood cancer
.
J Clin Endocrinol Metab
  .
2003
;
88
(
11
):
5307
5314
.
23.
Nicosia
SV
Matus-Ridley
M
Meadows
AT
.
Gonadal effects of cancer therapy in girls
.
Cancer
  .
1985
;
55
(
10
):
2364
2372
.
24.
Faddy
MJ
Gosden
RG
Gougeon
A
et al
Accelerated disappearance of ovarian follicles in mid-life: implications for forecasting menopause
.
Hum Reprod
  .
1992
;
7
(
10
):
1342
1346
.
25.
Wallace
WH
Thomson
AB
Saran
F
et al
Predicting age of ovarian failure after radiation to a field that includes the ovaries
.
Int J Radiat Oncol Biol Phys
  .
2005
;
62
(
3
):
738
744
.
26.
Coulam
CB
Adamson
SC
Annegers
JF
.
Incidence of premature ovarian failure
.
Obstet Gynecol
  .
1986
;
67
(
4
):
604
606
.
27.
Loren
AW
Mangu
PB
Beck
LN
et al
Fertility preservation for patients with cancer: American Society of Clinical Oncology clinical practice guideline update
.
J Clin Oncol
  .
2013
;
31
(
19
):
2500
2510
.
28.
The Ethics Committee of the American Society for Reproductive Medicine
.
Fertility preservation and reproduction in cancer patients
.
Fertil Steril
  .
2005
;
83
(
6
):
1622
1628
.
29.
British Fertility Society
.
A strategy for fertility services for survivors of childhood cancer
.
Hum Fertil (Camb)
  .
2003
;
6
(
2
):
A1
A39
.