Inclusion of heterozygotes for cystic fibrosis in the egg donor pool

Transcription

1 FERTILITY AND STERILITY VOL. 78, NO. 3, SEPTEMBER 2002 Copyright 2002 American Society for Reproductive Medicine Published by Elsevier Science Inc. Printed on acid-free paper in U.S.A. Inclusion of heterozygotes for cystic fibrosis in the egg donor pool Ulrike Zenke, M.D. and Ryszard J. Chetkowski, M.D. Alta Bates In Vitro Fertilization Program, Berkeley, California Objective: To describe and discuss our experience with cystic fibrosis (CF) carrier testing in a donor egg program. Design: Retrospective review. Setting: Community hospital-based assisted reproductive technology (ART) program. Patient(s): Forty anonymous white oocyte donor applicants. Intervention(s): Testing with a DNA mutation analysis panel. Main Outcome Measure(s): Frequency of heterozygotes for CF mutation among the donor applicants and the likelihood of carriers and noncarriers being selected by recipients. Result(s): Five of 40 egg donor applicants (12.5%) were found to be heterozygous for a CF mutation; 35 women (87.5%) tested negative. Two of the five CF carriers (40.0%) were selected by five recipient couples and underwent four donation cycles after the recipients male partners tested negative. Twenty-nine of the 35 noncarrier donors (82.9%) were matched and underwent 81 egg donation cycles. The likelihood of being selected was lower for CF carriers than for noncarriers. Conclusion(s): Our experience strongly supports the recommendation of routine CF testing of prospective white egg donors. Whereas heterozygosity lowers the probability of a donor being matched, it need not exclude her from the donor pool provided the recipient s partner is not a carrier. Empowering recipients to choose their own donors, focused patient education, and genetic counseling with precise determination of residual risk are important prerequisites for inclusion of CF carriers. (Fertil Steril 2002;78: by American Society for Reproductive Medicine.) Key Words: Carrier testing, cystic fibrosis, gamete donors, genetic screening, oocyte donation, residual risk Received October 22, 2001; revised and accepted February 8, Reprint requests: Ryszard J. Chetkowski, M.D., Alta Bates In Vitro Fertilization Program, 2999 Regent Street, Suite 101-A, Berkeley, California (FAX: ; abivf@pacbell.net) /02/$22.00 PII S (02)03248-X Cystic fibrosis (CF) is the most common autosomal recessive disease in the white population, yet most couples do not know whether they are at high risk because mass preconception screening has not been widely used. It was only in October 2001 that the American Colleges of Obstetricians and Gynecologists and Medical Genetics recommended offering CF screening to couples in whom one or both partners are Caucasian and are planning a pregnancy or seeking prenatal care even in the absence of family history (1). In whites of European extraction, the carrier frequency is 1/29 and the incidence of affected homozygotes is 1/3,300. The average life expectancy of homozygotes remains only 30 years with pulmonary complications being the primary cause of death. As in other autosomal recessive diseases, heterozygotes are generally healthy. The CF conductance regulator gene was identified in 1989 and more than 900 mutations and 70 DNA sequence variants have been described to date (2). A mutation in the CF conductance regulator gene causes defective chloride transport, leading to thick mucus secretions in the lungs, pancreas, sweat glands, and accessory male glands. A remarkable variability in the presence and severity of these manifestations has been observed, even when the same mutations are present (3). Obstructive azoospermia due to congenital absence of the vas deferens can be the sole clinical manifestation of the compound genotype, which includes two different mutant alleles, one of which is often uncommon and thus undetectable with most DNA screening assays (2). Screening programs for CF carriers must take into account the varying frequency of CF conductance regulator mutations in different populations, as well as the dependence of the 557

2 TABLE 1 The probability (residual risk) of being a carrier for CF after a negative test and the probability of having a child with cystic fibrosis when the egg donor is a known carrier and the recipient s partner tests negative by DNA mutation analysis in different populations. Ethnic group Risk of being a CF carrier before test Detection rate Risk of being a CF carrier after negative test Risk of offspring with CF after negative test Ashkenazi Jews a 1/29 97% 1/930 1/3,720 Northern Europeans b 1/29 90% 1/290 1/1,160 African Americans a 1/65 69% 1/207 1/828 Asian Americans b 1/90 30% 1/128 1/525 Southern Europeans b 1/29 74% 1/111 1/446 Hispanics a 1/46 57% 1/105 1/420 a Numbers are from the ACOG/ACGM guidelines (1). b Numbers are from the NIH Consensus Statement (4) and the Genzyme provider materials. Zenke. Cystic Fibrosis carriers in the egg donor pool. Fertil Steril detection rate on ethnic origin. With the now officially approved 25 DNA mutation panel, the detection rate ranges from 97% in Ashkenazi Jews to 30% in Asian-Americans, with an average of 80% in whites of European descent (1, 4). Consequently, the residual risk of having an affected offspring when one genetic parent is a known carrier and the other tests negative also varies widely from a low of 1/3,720 for Ashkenazi Jews to a high of 1/420 for Hispanic Americans (Table 1). In the past decade oocyte donation has become a common treatment for infertility but the screening of donors and the donor recipient matching process are not uniform. A recent survey of 159 assisted reproductive technology (ART) programs in the United States revealed that only 22% of centers tested for CF in the absence of family history (5). Traditionally, known carriers for CF and other autosomal recessive conditions have been excluded from the gamete donor pools (6, 7). However, the current guidelines of the American Society of Reproductive Medicine state that heterozygosity need not necessarily exclude a donor, but certain donors may be inappropriate in a given case (8). In view of the high variability in the residual risk of having an offspring with CF, we included CF carriers in the oocyte donor pool leaving the decision to educated recipients after full disclosure of the findings and calculation of their particular residual risk. This report describes our approach to the screening of donors and the egg donor recipient matching process, and defines the prerequisites that render inclusion of CF-positive donors medically appropriate. MATERIALS AND METHODS Between January 1998 and June 2001, 63 women took the initial written, telephone, and in-person interviews required of all prospective anonymous oocyte donors at the Alta Bates In Vitro Fertilization Program. The full evaluation consists of medical and genetic history, physical examination, vaginal ultrasound, tests for sexually transmitted diseases, baseline hormones, urine drug screen, and psychological evaluation including a structured interview and the Minnesota Multiphasic Personality Inventory-2. Carrier screening for the common autosomal recessive conditions is based on the donor s ethnic origin in accordance with the American Society of Reproductive Medicine guidelines (6, 8). The 40 fully or partly white donor applicants were tested for CF conductance regulator mutation by DNA probe analysis (Genzyme Genetics, Framingham, MA). When CF testing was introduced in our program for testing men with congenital absence of the vas deferens, there was as yet no officially sanctioned mutation panel. We chose the Genzyme assay because it was the most comprehensive. When the CF screening was extended to egg donors, we continued to use the same assay, although a more limited panel would have been acceptable. Now that American Colleges of Obstetricians and Gynecologists and Medical Genetics have defined an official 25-mutation panel, it seems advisable and less expensive to use this assay, except in groups where the detection rate is low. Initially, the Genzyme assay detected 70 of the most common mutations, but in May 2000 it was expanded to 86 mutations, without a significant change in the detection rate. According to the provider s literature, the detection rates of the assay are: 97% for Ashkenazi Jewish, 90% for Northern European, 75% for African-American, 30% for Asian-American, 57% for Hispanic, and 74% for Southern European populations. Compared to the now official 25-mutation panel, the detection rates are similar, except for a slightly higher sensitivity in African Americans. Depending on the result of the CF test, the subjects were divided into two groups: CF mutation positive and negative. 558 Zenke and Chetkowski Cystic fibrosis carriers in the egg donor pool Vol. 78, No. 3, September 2002

3 We compared the likelihood of being selected by prospective recipients for the two groups of donors. The study adhered to the guidelines of the Declaration of Helsinki. Because no clinical interventions were performed specifically for the purpose of this study and no subjects were contacted for interview, a separate application was not made to the Committee for the Protection of Human Subjects at the Alta Bates Medical Center. All of our patients consented to reporting of clinical results without identifying information. Statistical analysis was performed using the 2 test with significance defined as P.05. RESULTS None of the prospective donors reported a family history of CF. Five (12.5%) of the 40 prospective white egg donors were positive for a CF mutation and 35 (87.5%) tested negative. Three of the five CF-positive women had the F508 and one each the R117H and the I148T mutation. Two of the five CF-positive donors (40%) were selected by five different recipient couples and underwent four egg donation cycles. Twenty-nine of the 35 CF-negative donors (82.9%) were matched, a difference that was statistically significant (P.03). One of the heterozygous donors underwent three egg donation cycles and neither of the two children born to date have manifested CF. The second CF-positive donor was selected by two recipient couples, but the first match had to be canceled because the recipient s partner also tested positive for a CF mutation. The ethnicity of the recipients husbands was: one Ashkenazi Jewish, three Northern, and one Southern European. Both of the CF carrier donors selected possessed educational, personal, and physical characteristics in high demand among our recipients. The couples who chose CF carriers were highly educated and demonstrated excellent understanding of the potential risks. Another donor who tested positive for a CF mutation had previously participated in egg donation elsewhere in California without having been tested for CF. DISCUSSION Traditionally, screening for the common autosomal recessive conditions has been used to exclude carriers from the gamete donor pool. The 1990 American Society of Reproductive Medicine guidelines stated that a donor should not carry an autosomal recessive gene for any disease known to be prevalent in the donor s ethnic background for which heterozygosity can be detected but the rationale for this recommendation was not provided (6). Thus, the 1998 American Society of Reproductive Medicine guidelines, which permit inclusion of heterozygous donors, represent a major break with long-standing tradition and reflect a new perspective on genetic testing and inherited diseases (8). In our small sample of 40 white egg donors, one of eight was found to be a carrier for CF, a finding that strongly supports the American Society of Reproductive Medicine recommendation of routine testing. Because heterozygosity reduced a donor s chance of being matched by more than half, CF testing is best done at the outset of the medical evaluation. Thus, the applicant can be informed that she is less likely to be selected, she can be offered genetic counseling and an opportunity to withdraw from further testing for which there is no compensation in our program. In addition, certain low yield tests, such as human immunodeficiency virus (HIV) and human T-lymphotropic virus (HTLV) antibodies, which are required by the California Health and Safety Code, but have been invariably negative in our donor population, can be deferred until she has been matched, which may take many months. To our knowledge, ours is the first report of intentional incorporation of CF carriers into a donor egg program. Therefore, it behooves us to justify this novel practice by carefully considering and refuting the reasons why most ART programs have followed a diametrically different policy. Clearly, the well-being of the children about to be conceived is paramount. The primary reason why heterozygotes for CF have been excluded from the gamete donor pool is the unacceptably high risk (1/4) of producing an offspring with a lethal condition if the recipient s partner is also a carrier. All of us live with a few rare mutations that would be lethal in the homozygous state but the chance of sharing such abnormal alleles is infinitesimally small. Hence, the mere fact of heterozygosity for some mutation cannot be a rational basis for exclusion. Even when the recipient s partner tests negative for CF, there remains a residual risk due to the fact that the detection rate is less than 100%. However, for all but one ethnic group listed in Table 1, the residual risk of having a child with CF is lower than the 1/435 risk of a chromosomal abnormality with eggs of a 28-year-old woman, which happens to be the average age of donors in our program. The risk of Down s syndrome with a 28-year-old woman s eggs is 1/1,053. Thus, from the viewpoint of pure probabilities, the magnitude of this risk falls within the range that most recipients find acceptable for other serious conditions; a high percentage of our recipients do not undergo prenatal genetic testing for the detection of chromosomal anomalies. Clearly the patients must understand that, unlike aneuploidy, fetuses homozygous for CF cannot be detected by amniocentesis if the same DNA assay that missed the male partner s mutation is used. With ever-increasing ethnic diversity, difficulty may be encountered in quantifying the residual risk facing a couple. This uncertainty has been given as another reason for excluding heterozygotes (9). Within the past few months, the technique of sequencing the entire CF gene has become commercially available (Ambry Genetics, Costa Mesa, CA). This direct mutation analysis detects all the known exons and their surrounding regions on the CF conductance regu- FERTILITY & STERILITY 559

4 lator gene as well as the common intronic variants. Gene sequencing identifies more than 90% of the mutations responsible for CF in the white, Hispanic, and African-American populations, thus lowering the residual risk to less than 1/1,100 in these ethnic groups. At this time there is insufficient data on the CF mutation frequency and detection rate in the Asian Americans. Gene sequencing should identify most homozygous fetuses, even when one of the genetic parent s DNA mutation analysis is falsely negative. The third reason why carriers for autosomal recessive conditions are excluded from gamete donation is to avoid transmission of heritable genetic disorders as half of the children would also be heterozygous (7). This argument follows an old eugenic tradition that aims to improve a population by selecting only its best specimens for breeding. We certainly do not subscribe to the notion that ART program policies should strive to improve the gene pool. Like most clinics, Alta Bates has long accepted men with congenital absence of the vas deferens, who have two different CF mutations, for treatment by intracytoplasmic sperm injection (ICSI) if their spouses test negative. Numerically, excluding every autosomal recessive carrier from donating eggs would have negligible effect on the gene pool because all the donor egg deliveries represent only about 0.005% of the approximately 3.5 million annual births in the United States. Most heterozygotes for CF are healthy despite apparent over-representation among patients with asthma and severe oligoasthenospermia without congenital absence of the vas deferens (2, 10). The limited detection rate of DNA assays makes the diagnosis of heterozygosity presumptive rather than definitive because compound genotype, including a common and a rare mutation, cannot be ruled out. Contrary to common belief, infertile men with congenital absence of the vas deferens are not true heterozygotes but have two different mutant alleles, one or both of which are more uncommon and mild than those responsible for CF (2). In discussing their policy of excluding CF and other autosomal recessive carriers, Wallerstein et al. (9) cited reluctance to create a child who has a 50% chance of being a carrier but they provide no data to support their assertion. Our experience shows that, whereas recipients are less likely to select an egg donor who is a carrier for CF, some couples do freely choose heterozygotes. Thus a blanket policy excluding all autosomal recessive carriers from the donor pool reflects reluctance on the part of the treating physicians when properly the decision belongs to the recipient couple. Because the future child s well-being must be the touchstone of policy and the would-be parents are the best guardians of their child s welfare, our policy of leaving the decision to the parents, rather than medical experts, accords with the universal postnatal assignment of such responsibility to parents as well as the existing research on the societal perception of risk (11). Assessment of risks by the public seldom coincides exactly with expert opinion, but the experts in the field of risk assessment wisely conclude that in our democracy, people are final arbiters of how safe is safe enough (12). Ultimately, one cannot consider the issue of how CF carriers are managed in donor egg programs without reference to the values underlying the different approaches to the egg donor recipient matching process. Many clinics follow a paternalistic paradigm and assign donors to recipients, whereas other programs, such as ours, are libertarian and expect recipients to choose their own donors. The former programs justify their practice by invoking the ethical principle of beneficence, whereas the latter consider the patient s autonomy paramount. Comparison of these two models reveals a major difference in the assumption of responsibility by either the physician or the patient, respectively. Fear of liability may be the real reason why clinics are reluctant to accept CF carriers as donors. Why have mutations resulting in a lethal disease and male infertility become so common? It is likely that heterozygosity for CF confers an evolutionary advantage the nature of which has not been elucidated to date (13, 14). In the case of sickle cell anemia, carriers have long been known to be protected from malaria. Are there any advantages to including CF heterozygotes among egg donors? Discussing CF makes patients aware of the complexities inherent in the screening and selection of donors, thus raising the informed consent process to a higher level. We have been repeatedly struck by how many highly educated consumers think that it is possible to test for everything and how few have any appreciation for the gray zones so ubiquitous in medicine. Last but not least, most recipients desire as wide a choice of donors as possible and inclusion of CF carriers can increase the number of candidates to consider maximizing the recipients degrees of freedom. In our program, the recipients participate in at least one psychoeducational session before face-to-face discussion of donor selection with their physician. It is their treating physician who makes a critical assessment of whether a given couple is in a good position to deal with all the ramifications of CF. Language barrier, limited educational background, and unusually high anxiety level may make it inadvisable to discuss CF-positive donors. However, most recipients are deemed fit and receive detailed medical and genetic information about CF from their physician before being asked whether or not they wish to consider such donors. In our opinion it is important that the decision to consider CF carriers be made in the abstract before any specific donors are discussed so as to avoid any inducement to undue risks. The majority of recipients elect to look only at noncarriers. Those who do wish to consider CF carriers also have several CF-negative donors to choose from. All patients have 560 Zenke and Chetkowski Cystic fibrosis carriers in the egg donor pool Vol. 78, No. 3, September 2002

5 ample opportunity to discuss genetic and other issues with their doctor and are encouraged to meet with a genetic counselor. Whereas empowering patients to choose their gamete donors appears necessary for the inclusion of heterozygotes, it is not in itself sufficient. Sperm banks offer their clients free choice of donors but the process typically occurs without any medical or genetic guidance. In the absence of focused patient education and counseling, consumers cannot be expected to deal with an issue as complex as CF on their own. Hence, it is appropriate for sperm banks to continue their policy of excluding autosomal recessive carriers. For the couples who choose to consider CF carriers, heterozygosity is just one of many characteristics they evaluate in a donor. Recipients have diverse and frequently inchoate criteria and priorities, and their final decision often appears to occur at the gut level. The egg donor selection process is inevitably complex and many people find it awkward, difficult, and stressful. It is the physician s responsibility to provide his or her patients with the tools they require to best exercise their autonomy in making this most personal and weighty choice. The approach we have developed at Alta Bates is designed to help recipients integrate the entire gamut of medical, genetic, educational, ethnic, physical, psychological, and personal issues in a nonjudgmental way. References 1. The American College of Obstetricians and Gynecologists and American College of Medical Genetics. Preconception and prenatal carrier screening for cystic fibrosis: clinical and laboratory guidelines. October Lewis-Jones DI, Gazvani MR, Mountford R. Cystic fibrosis in infertility: screening before assisted reproduction. Hum Reprod 2000;15: Kuller JA, Baughman R, Biolsi C. Cystic fibrosis and the National Institutes of Health consensus statement: are obstetrician gynecologists ready to comply? Obstet Gynecol 1999;93: Genetic testing for cystic fibrosis. National Institutes of Health consensus statement. 1997;15: Lewis V, Saller DN Jr, Kouides RW, Garza J. Survey of genetic screening for oocyte donors. Fertil Steril 1999;71: American Fertility Society. Ethical considerations of the new reproductive technologies: appendix B: minimal genetic screen for gamete donors. Fertil Steril 1990;53(Suppl 2):88S 9S. 7. ACOG committee opinion. Genetic screening of gamete donors. Int J Gynaecol Obstet 1998;60: American Society for Reproductive Medicine. Guidelines for gamete and embryo donation: appendix A: minimal genetic screening for gamete donors. Fertil Steril 1998;70(Suppl 3):12S 3S. 9. Wallerstein R, Jansen V, Grifo JA, Berkeley AS, Noyes N, Licciardi F, et al. Genetic screening of prospective oocyte donors. Fertil Steril 1998;70: Dahl M, Tybjaerg-Hansen A, Wittrup HH, Lange P, Nordestgaard BG. Cystic fibrosis Delta F508 heterozygotes, smoking, and reproduction: studies of 9141 individuals from a general population sample. Genomics 1998;50: Slovic P. Perception of risk. Science 1987;236: Lave LB. Health and safety risk analyses: information for better decisions. Science 1987;236: Jorde LB, Lathrop GM. A test of the heterozygote advantage hypothesis in cystic fibrosis carrier. Am J Hum Genet 1998;42: Meindl RS. Hypothesis: a selective advantage for cystic fibrosis heterozygotes. Am J Phys Anthropol 1987;74: FERTILITY & STERILITY 561