The latest official guidelines on cystic fibrosis screening have some clinicians bewildered--and others looking for an easy way to put them to good use. A top expert in the field provides practical advice on how to individualize the recommendations.
The latest official guidelines on cystic fibrosis screening have some clinicians bewilderedand others looking for an easy way to put them to good use. A top expert in the field provides practical advice on how to individualize the recommendations.
The viscous secretions that accumulate in the pulmonary and pancreatic ducts of patients with cystic fibrosis (CF) can cause chronic obstructive lung disease, pancreatic enzyme deficiencies, and all the clinical complications that accompany these two defects.
The disease's pathogenesis centers around a defect in a chloride channel that is essential for excreting chloride and inhibiting sodium uptake in epithelial cells (Figure 1). This common autosomal-recessive disorder occurs in all ethnic groups but is especially prevalent in Caucasians of Ashkenazi Jewish or European ancestry, affecting one in 3,300; one in 27 is a heterozygotic carrier.1,2 The disease occurs much less frequently in African-Americans, Asians, and Hispanics.1,2 Over 30,000 Americans have severe CF, and 8 to 9 million are heterozygotes.
PLEASE NOTE: If you are using Internet Explorer version 6 or higher, after clicking to enlarge the graphic, click the picture icon that appears in the lower right-hand corner of the graphic.
Because of CF's high prevalence and severity, The American College of Obstetricians and Gynecologists now recommends genetic screening for all asymptomatic women. Guidelines promulgated in 2001 have led ob/gyns to develop individual office procedures to help implement these recommendations.3 My purpose here is to review the guidelines and their rationale, and to suggest ways to apply these in the real world of clinical practice.
Although CF is a severe disease, not all affected individuals show pulmonary and pancreatic manifestations. Mean life expectancy is 30 to 35 years, and has increased only slightly over the last few decades.4 In the United States, a significant portion of lung transplants and heart/lung transplants are now done on patients with the disease. The disorder usually produces clinical signs and symptoms in early childhood, but 10% to 20% of affected individuals are detected at birth because of meconium ileus.
Increasing accumulation of viscous secretions can gradually bring on chronic respiratory obstruction, while blocked pancreatic ducts and the subsequent poor intestinal absorption often cause malnutrition and postnatal growth retardation. Almost all males with CF have azoospermia, the anatomic result of congenital bilateral absence of the vas deferens. The "gold standard" for diagnosis remains the chloride sweat test, although clinicians increasingly use molecular confirmation as well. Once one of the mutations discussed further on is recognized in a family, molecular studies can detect heterozygotes and affected relatives.
The CF geneofficially called the Cystic Fibrosis Transmembrane Regulator (CFTR)was cloned in 1989.5 Located on the long arm of chromosome 7, it consists of 27 exons. Given its large size, opportunities exist for many mutations to develop, all of which could ultimately result in a dysfunctional or absent gene product; about 1,000 different mutations have already been recognized. Molecular tests can readily detect a specific mutation, and willing vendors can sequence the entire gene. Keep in mind, however, that CFTR mutations will still not be found in all affected or heterozygous individuals. A test's sensitivity for finding a mutation in a known heterozygote varies by ethnic background.
In most populations, the most common mutation involves a deletion (
) of the three nucleotides at codon 508, resulting in the loss of the amino acid phenylalanine (F)
F508 (Figure 2). The most common defect in Ashkenazi Jews involves a nucleotide mutation in the codon that usually results in the amino acid tryptophan (W). In this mutation, a stop codon (X) is inappropriately present at position 1282W1282X, resulting in the absence of a functional CFTR gene product. Among the many other mutations, only 20 to 25 occur as frequently as 1 per 1,000 (0.1%) in the general population.1
Given the high prevalence of CF, screening seemed likely to be offered soon after the gene was cloned in 1989. But the situation immediately proved more complicated than was the case with sickle cell anemia, the prototypic molecular disorder. In the latter case, virtually all mutations are caused by a single point mutation (codon 6), which results in the hemoglobin sickling. In contrast, CF is characterized by molecular heterogeneity, as witnessed by the 1,000 known mutations. Although
F508 was recognized as the most common mutation, the extent to which it contributes to the clinical syndrome varies widely from population to population. In Caucasians of northern European descent, including the British, 70% or more of all CF mutations are due to
F508; in Hispanics, however,
F508 is identified in only 50% to 55%, and in some Slavic populations barely 25%.
This heterogeneity means that population screening would be inefficient if we only tested for
F508. It would only detect about 60% of CF chromosomes, in heterozygous or homozygous form. Screening only for
F508 would leave many couples in the anxious position of knowing that one partner is heterozygous, but unable to exclude the other as a carrier. Worse yet, fetal diagnosis would not be possible through amniocentesis or chorionic villi analysis because there is no biochemical assay for the protein gene product.
Plans for CF screening were thus put on hold pending molecular advances and studies assessing receptivity of patients. On the molecular front, work began that eventually led to identification of some 1,000 CF mutations. Most are rare, but aggregate screening for 23 specific mutations can detect 97% of CF mutations in Ashkenazi Jews, 80% in Caucasians of European ancestry, and lower frequencies in Hispanic, Asian, and African-Americans (Table 1). Researchers have shown that CF screening could be performed without undue alarm, and that patients want it. The positive experience with Tay-Sachs disease should apply equally to CF. The consensus developed that once 80% or more of CF cases could be detected (vs. 60% with
F508 CF testing alone), population screening would become attractive.
In 1997 an NIH consensus conference considered laboratory, epidemiologic, and sociologic data and finally recommended population screening for cystic fibrosis.6,7 Yet, the conference provided no specific instructions on how to carry through on this recommendation. Nonetheless, the report spurred development of a task force consisting of representatives from the NIH Human Genome Center, ACOG, and the American College of Medical Genetics (ACMG). The task force developed guidelines, created patient brochures, and generated public education materials. They recommended a panel of 25 mutations for screening, including
F508, W1282X, and 23 others.1 In 2004, two of these mutations were dropped, leaving the current recommendation at 23 mutations.
The group said CF screening should be offered to Caucasians of Ashkenazi Jewish or European descent while it should be made available in other ethnic groups. Distinctions between "made available" and "offered" were not provided.
Once these guidelines were introduced in 2001,2,3,8 ob/gyns began introducing CF screening in their own practices. CF thus joined other genetic disorders for which ACOG recommended population screeningTay-Sachs disease and Canavan disease in Ashkenazi Jews, sickle cell anemia in African-Americans, ß-thalassemia in individuals of Southern Mediterranean heritage (e.g., Greek, Italian, Cypriot), and
-thalassemia in certain Asians (e.g., Vietnamese, Chinese, Filipino).9
Of course the sudden introduction of any screening policy can generate confusion and unexpected consequences, so gradual introduction of CF screening was considered wise. By 2004, practices had the opportunity to develop their own tailored protocol. Screening ideally should occur before pregnancy, but typically takes place postconception. This should occur early enough to allow patients time for various reproductive options. One must first exclude couples with a positive family history, for CF screening as described here is applicable only to those with a negative family history. (If a relative has CF, obtain records or refer the couple to a geneticist who can determine the molecular basis. His or her recommendations are likely to involve looking for that particular mutation in your patient or her husband. The partner with the negative family history can be screened with a standard CF panel.)
Selecting the test panel. Fulfilling ACOG/ACMG guidelines for the 23 mutations is possible through several vendors. Costs typically are $150 to $300. Some offer larger panels, screening 35 to over 70 mutations. Panels that screen for more than the 23 mutations aren't necessary if there's a negative family history: Few additional heterozygotes are detected beyond the one in 27 expected in Caucasians.
However, using an extended panel is not unreasonable. If that's the approach you choose, it's best to manage every patient in your practice similarly. Similar reasoning applies to so-called ethnic-specific panels. Again, detection rates are currently only marginally improved over the standard CF panel, but ethnic-specific panels may be valuable if a positive family history exists.
Deciding to screen one or both partners. It is advisable that every practice develop a protocol for CF screening. No single protocol is universally applicable, but consistency is crucial. One way to document compliance is to develop a written protocol to which all staff adhere. A key decision is whether to simultaneously screen the prospective mother and father or screen only the mother and then the father if the former is a carrier (the sequential approach). Either approach is acceptable. Screening both parents obviously produces the highest detection rate, although not by much (Table 2). The downside is that one partner will be a carrier twice as often, generating more anxiety and requiring follow-up more often. Keep in mind, though, that in both concurrent and sequential screening, you still won't detect every CF mutation.
Alerting relatives. Suppose a patient or her spouse has a severe CF-causing mutation like
F508. Relatives of your patient could well have inherited the same mutation from a common ancestor. The likelihood is 50% that any one of your patient's siblings or her partner's siblings will have the same mutation. As caring physicians, we want to alert at-risk individuals; however, we have no authority to contact a relative directly. The solution is using a sample letter, mailed not by the physician but by the patient or her spouse. This letter would inform relatives of the consequence of a CF mutation having been detected. A template for such a letter exists in the ACOG/ACMG booklet, Preconception and Prenatal Carrier Screening for Cystic Fibrosis.2
"Postmarketing" surveillance. Does CF screening perform as predicted? Yes. One large study involving Kaiser Permanente California screened 27,000 women.10 The heterozygote frequency expected was 1 in 28; 1,000 affected fetuses were detected. The costs of screening equaled the projected cost of caring for affected liveborns, which were avoided by patients exercising reproductive options.
"Offering" the test panel vs. "making it available." Which patients should be screened? ACOG/ACMG guidelines state that screening should be offered to Caucasians of European and Ashkenazi Jewish ancestry. There is no further instruction, but it is reasonable to assume that the approach should be similar to that used in Tay-Sachs disease, sickle cell anemia, or ß- and
-thalassemia. But first, clinicians must start a dialogue with their patients. Remember that genetic screening is never obligatory, as it might be for Rh(D) screening. A patient brochure produced by ACOG can help you review the inheritance of CF, variability of CF symptoms, and likelihood of detecting carrier status. This brochure explicitly states that not all CF mutations are detectable, and, hence, not all affected individuals can be detected. In other words, false-negative cases are unavoidable. We can't overemphasize the legal significance of this statement.
In couples of African-American, Hispanic, and Asian origin, CF screening should be "made available." Unfortunately the distinction between "offering" and "made available" isn't very clear. Perhaps a brief informative statement about CF could be followed by explaining to the patient that CF screening exists, and that information through brochures or other sources can help her decide if she should pursue it further. The ACOG/ACMG brochure, Cystic Fibrosis Carrier Testing: The Decision is Yours, can be helpful. Some physicians find that it is unwieldy to handle patients of different ethnicities within a given practice. In many multiethnic practices it may be difficult to stratify patients by ethnic group.
Documenting compliance. How should you document compliance with the guidelines? Adhere to a consistent plan of verifying that the screening information has been conveyed. This might consist of a written note in the chart, or retention of written informed consent stating the patient's choice. A sample consent form exists in the ACOG patient information brochure mentioned previously; however, retaining such a form on every patient might be burdensome, especially if an office uses an electronic medical record.
Follow-up after CF screening. If CF tests on both partners are negative, no further evaluation is required. Be certain each couple realizes that 100% carrier detection is never possible in general screening (Table 1). If, on the other hand, one partner has a mutation, but the other screens negative, no further evaluation is needed. The residual risk of still having a CF fetus is low but finite (Table 2).
Genetic counseling is not ordinarily required, but some patients may request it. If one partner proves to be a CF carrier and the other is not tested, the risk is not inconsequential. In Europeans of Caucasian origin, the residual risk is one in 116. One approach is chorionic villi or amniotic fluid analysis to exclude the known mutation.
If both couples are heterozygous for the same or for two different deleterious alleles (compound heterozygotes), refer them to a geneticist. The risk of an affected fetus is 25%, and options for a definitive diagnosis should be presented.
The ACOG/ACMG booklet, Cystic Fibrosis Testing: What Happens if Both my Partner and I are Carriers? can be useful at this stage of screening.11
5T polymorphism and CBAVD. In intron 8 (the noncoding regions between exons 8 and 9) a polymorphism exists that's relevant to CF. (A polymorphism is an alternate form of a gene, examples being blood groups A, B, or O, or the various HLA alleles.) The CF polymorphism involves the presence of either 5, 7, or 9 thymidines (5T, 7T, 9T) at a certain position. If the 5T allele is present, splicing the noncoding intron from the coding exon becomes inefficient on that chromosome. With 5T, the amount of CF proteinthe gene productis approximately 50% of normal. By contrast, in
F508 and W1282X virtually no CFTR gene product is synthesized. In other words, 5T by itself is not sufficiently harmful to produce severe CF, with pulmonary and pancreatic disease. But 5T may cause congenital bilateral absence of the vas deferens (CBAVD) if present in homozygous form, or if present in compound heterozygosity along with a more severe CFTR mutation like
If the sole purpose of CF in population screening were to exclude pancreatic and pulmonary disease, testing for the 5T polymorphism would not be required. The rationale is that CBAVD is not a severe form of CF. However, if 5T exists on the same chromosome (i.e., cis) as a particular second mild mutation (specifically the point mutation R117H), the combination becomes comparable to a severe CF mutant allele like
F508; very little CFTR is synthesized:
If 5T exists on the same chromosome as R117H and the other chromosome has another severe mutation (
F508), severe CF will exist.
If 7T is present on the same chromosome as R117H, the fetus is only at risk for mild CF, even if the allele on the other chromosome is
F508. CBAVD will result, but not severe CF.
If both chromosomes show 5T (homozygosity), and there is no other CF mutation, CBAVD can exist but not severe CF.
In consideration of the above, labs routinely test (reflex) for 5T whenever R117H is detected. Otherwise, 5T will not be sought. If a physician wishes to know the status of 5T polymorphisms, for example in a workup for male infertility, he or she must specifically request it.
In summary, physicians should offer CF screening to Caucasians of European or Ashkenazi Jewish ancestry and make the test available to other ethnic groups. Although it's best to screen a nonpregnant patient, it usually doesn't happen that early on. A specific panel of mutations is now recommended by ACOG/ACMG, encompassing the relatively common
F508 and W1282X alleles and 21 less common mutations. This panel will detect most, but importantly not all, carriers and, hence, affected fetuses. You have to make patients realize that the screening test will not detect all carriers. Patient information brochures may help you implement guidelines, and a standard protocol should be followed in offices.
Dr. Simpson is on the scientific advisory board for Imetrikus.
1. Grody WW, Cutting GR, Klinger KW, et al. Laboratory standards and guidelines for population-based cystic fibrosis carrier screening. Genet Med. 2001;3:149-154.
2. ACOG, ACMG. Preconception and Prenatal Carrier Screening for Cystic Fibrosis. Washington DC; The American College of Obstetricians and Gynecologists, 2001.
3. Mennuti MT. Lights! Camera! Action! Obstet Gynecol. 2001;98:539-541.
4. Cutting GR. Cystic fibrosis. In: Rimoin DL, Connor JM, Pyeritz RE, Korf B, eds. Principles and Practices of Medical Genetics. Edinburgh: Churchill-Livingston; 2002:1561-1606.
5. Riordan JR, Rommens JM, Kerem B, et al. Identification of the cystic fibrosis gene: cloning and characterization of complementary DNA. Science. 1989;245:1066-1073.
6. NIH Consensus Development Conference Statement genetic testing for cystic fibrosis. Genetic testing for cystic fibrosis. Arch Intern Med. 1999;159:1529-1539.
7. Mennuti MT, Thomson E, Press N. Screening for cystic fibrosis. Obstet Gynecol. 1999;93:456-461.
8. Cystic Fibrosis Foundation: Patient Registry 1999 Annual Report, Bethesda, MD, September, 2000.
9. Simpson JL, Elias S. Genetics in Obstetrics and Gynecology. 3rd ed. Philadelphia, Pa: Saunders; 2003:83-98.
10. Witt DR, Coppinger J. Cystic fibrosis prenatal screening of 27,000 women in a large HMO. Am J Hum Genet. 2000;69(4):176.
11. ACOG. The Decision is Yours, booklet #10. Cystic Fibrosis TestingWhat Happens if Both my Partner and I are Carriers? 2001.
Joe Leigh Simpson. Making sense of the CF screening guidelines. Contemporary Ob/Gyn Oct. 1, 2004;49:60.