American Journal of Medical Genetics 79:175–183 (1998)

Marker Segregation Information in Breast/Ovarian

Cancer Genetic Counseling: Is It Still Useful?
Laurent Essioux,1 Catherine Girodet,2 Olga Sinilnikova,3 Sabine Pagès,4 François Eisinger,5
Sandrina de Résende,6 Christine Maugard,7 Didier Lanoë,7 Michel Longy,6 Yves-Jean Bignon,2
Hagay Sobol,5 Catherine Bonaı̈ti-Pellié,1 Dominique Stoppa-Lyonnet,4* and the Groupe Génétique
et Cancer de la Fédération Nationale des Centres de Lutte Contre le Cancer
1
Unité de Recherches en Epidémiologie des Cancers/INSERM U351, Institut Gustave Roussy, Villejuif, France
2
Unité d’Oncogénétique/INSERM CRI9402, Centre Jean Perrin, Clermont-Ferrand, France
3
Centre International de Recherche sur le Cancer, Lyon, France
4
Unité de Génétique Oncologique, Institut Curie, Paris, France
5
Département d’Oncologie Génétique/INSERM CRI9703, Institut Paoli-Calmettes, Marseille, France
6
Laboratoire d’Oncologie Moléculaire, Institut Bergonié, Bordeaux, France
7
Oncologie Génétique, Centre René Gauducheau, Site Hospitalier Nord de Nantes, Saint-Herblain, France

The use of mutation screening of BRCA1
and BRCA2 genes as a genetic test is still to
a certain extent limited and the oncogeneti-
cist may want to use complementary ap-
proaches to identify at-risk individuals. In a
series of 23 families with at least three
breast or ovarian cancer cases, screened for
mutations at BRCA1 and BRCA2 and typed
for markers at both loci, we investigated the
usefulness of marker segregation informa-
tion at two levels: 1) to what extent can the
indirect approach identify the mutation
carrier status of screened cases and their
first-degree relatives, and 2) in what way
does it help to identify the gene implicated
in a family in which neither BRCA1 nor
BRCA2 mutation has been detected? Using
the indirect approach, the carrier status of
the screened case could be determined with
Contract grant sponsor: Axe Génétique et Cancer de la Ligue
Nationale Contre le Cancer (LNCC); Contract grant sponsor: the
Comités Départementaux of Puy de Dôme, Bouches-du Rhône,
Alpes de Hautes-Provence, Var, and Val de Marne of the LNCC;
Contract grant sponsor: Association pour la Recherche sur le Can-
cer (ARC); Contract grant sponsor: Programme Hospitalier de
Recherche Clinique; Contract grant number: IDF94014; Contract
grant sponsor: Fondation de France; Contract grant sponsor:
Caisse Nationale d’Assurance Maladie; Contract grant sponsor:
CAMPLP; Contract grant sponsor: GREG; Contract grant spon-
sor: Association Charbonnières Triathlon; Contract grant spon-
sor: FEGEFLUC; Contract grant sponsor: INSERM 4U003C;
Contract grant sponsor: IFSBM (Institut Supérieur de Formation
aux Sciences BioMédicales); Contract grant sponsor: Comité de
l’Ain of the LNCC.
*Correspondence to: Dominique Stoppa-Lyonnet, Unité de Gé-
nétique Oncologique, Institut Curie, 26 rue d’Ulm, 75231 Paris
Cedex 05, France. E-mail: Dominique.lyonnet@curie.fr
Received 14 October 1997; Accepted 6 April 1998
© 1998 Wiley-Liss, Inc.
quasi certainty in three families and with a
high probability in eight families. This sta-
tus could be inferred in unaffected first-
degree relatives as almost certain in one
family and as highly probable in six fami-
lies. Fourteen mutations were found con-
currently in our series. Among the nine mu-
tation-negative families, we were able to
conclude that a BRCA1 mutation most prob-
ably segregated in one and that a mutation
other than BRCA1 and BRCA2 was probably
involved in two families. Our results show
that, in small families, little help is to be ex-
pected from linkage data and mutation
screening is the only way of identifying the
origin of a genetic predisposition in a fam-
ily. Marker segregation information may be
useful in some large breast/ovarian cancer
families in which no BRCA1 or BRCA2 mu-
tation has been detected. Am. J. Med. Genet.
79:175–183, 1998. © 1998 Wiley-Liss, Inc.
KEY WORDS: genetic counseling; breast/
ovarian cancer; genetic pre-
disposition; marker segrega-
tion information; BRCA1;
BRCA2
INTRODUCTION
The identification of BRCA1 [Miki et al., 1994] and of
BRCA2 [Wooster et al., 1995; Tatvigian et al., 1996]
were major events in the understanding of predisposi-
tion to breast and ovarian cancer. Since then, 452 dif-
ferent BRCA1 mutations and 254 different BRCA2 mu-
tations were registered by January 1998 in the BIC
database (available on the World Wide Web at http://
176 Essioux et al.
www.nhgri.nih.gov/Intramural research/Lab transfer/
Bic). They include nonsense mutations, frameshift de-
letions or insertions, missense mutations, and splicing
aberrations found throughout these genes.
Mutation screening as a genetic test is to a certain
extent limited. In order to reduce the false negative
rate, the mutation identification strategy adopted must
involve different detection methods to detect these dif-
ferent types of mutations [Friedman et al., 1995; Se-
rova et al., 1996; Puget et al., 1997]. In addition, Couch
et al. [1996] have reported 15% of BRCA1 variants with
an unknown biological significance, which raises the
problem of risk assignment to putative carriers. The
report on a BRCA2 mutation that should result in a
truncated (and therefore nonfunctional) protein, with-
out clear association with the disease status, docu-
ments the difficulty of ascertaining the deleterious ef-
fect of a mutation [Mazoyer et al., 1996].
Alternatively, joint segregation of genetic markers
with the disease may be used to infer the genetic status
of relatives in a family. This approach, known as the
indirect approach, is useful when genes are localized
but not yet cloned. Even when genes are cloned, this
approach may be used when mutations are not always
detected. For instance, Spirio et al. [1993] developed a
genetic test with a polymorphic intragenic marker in
familial adenomatous polyposis (FAP), which was
shown to be efficient and a reliable alternative in fami-
lies where no mutation of the APC gene is found.
Breast and ovarian cancer predisposition is more com-
plex than FAP, due to heterogeneity, incomplete pen-
etrance, and a high frequency of phenocopies, leading
to uncertain inference on genotypes from phenotypes.
Before the cloning of BRCA1, Rowell et al. [1994]
stated that genetic counseling using linkage informa-
tion was only possible in families with a large number
of cases and a convincing evidence of linkage to
BRCA1. However, this requirement might be less man-
datory if prior probabilities of linkage to BRCA1 and
BRCA2, estimated by the Breast Cancer Linkage Con-
sortium (BCLC) [Ford et al., 1998], and data on the
segregation of markers at these loci are combined.
Besides being a tool for genetic testing, marker seg-
regation at these loci may help to pinpoint which gene
segregates in a family in which no mutation is found
and thus explain the reason for such a result (the
screened case was a phenocopy, the mutation was
missed, or another gene probably segregates in the
family). In their study on genetic heterogeneity, Reb-
beck et al. [1996] attributed their familial observations
to BRCA1 or BRCA2 using two criteria: a positive mu-
tation screening result or, in the mutation-negative
families, suggestive evidence of linkage to one of these
two genes. More recently, Serova et al. [1997] calcu-
lated the probability that aggregation of cases in a fam-
ily is due to BRCA1, BRCA2 or BRCAx (any other un-
known gene) by combining linkage information and
negative mutation screening results.
In a series of 23 families selected from French cancer
genetics clinics, we investigated the usefulness of
marker segregation information at two levels: 1) to
what extent can the indirect approach identify the mu-
tation carrier status of screened cases and their first
degree relatives, and 2) in what way does it help to
identify the gene implicated in a family in which nei-
ther BRCA1 nor BRCA2 mutation has been detected?
SUBJECTS AND METHODS
Family Sample
Twenty-three families with at least three cases of
breast or ovarian cancer among first- or second-degree
relatives in the same parental lineage, were ascer-
tained from five French cancer genetics clinics of the
French Cooperative Network (Groupe Génétique et
Cancer de la Fédération Nationale des Centres de
Lutte Contre le Cancer). They were selected for genetic
counseling with the indirect approach because they
had at least three members available for DNA typing.
Table I gives a description of the family sample.
Individuals Tested for Mutation Screening or
DNA Marker Typing
In each family, BRCA1 and BRCA2 mutations were
screened in one individual, selected on the basis of a
low probability of being a sporadic case, usually a
woman affected with either an early-onset breast can-
cer or an ovarian carcinoma. In one family, the
screened case was an affected man, and in another
family, two cases were screened.
The efficiency of the indirect approach was assessed
by evaluating its ability to determine the mutation car-
rier status in affected and in unaffected individuals.
Risks of being a BRCA1 or a BRCA2 mutation carrier
using marker segregation (see ‘‘Risk Calculations’’ be-
low) were calculated in each family for the screened
case and for a young unaffected female (or an unaf-
fected male) first-degree relative of the screened case.
The prior probability that these individuals are muta-
tion carriers is close to 0.5, provided that the screened
case is a mutation carrier.
Analysis of BRCA1 and BRCA2 Loci
Polymorphic Markers
DNA was extracted from either whole blood or Ep-
stein-Barr virus–immortalized lymphoblastoid cell
lines according to standard methods [Sambrook et al.,
1989]. Families were genotyped with four polymorphic
microsatellite markers flanking BRCA1, D17S579/
mfd188 and D17S800/AFM200zf4, and BRCA2,
D13S260 and D13S267 [Hall et al., 1990; Weissenbach
et al., 1992; Wooster et al., 1994]. Polymerase chain
reaction (PCR) was performed using 50 to 100 ng of
genomic DNA, in a 20-␮l reaction containing MgCl2 1.5
mM, dNTPs 100 ␮M each, primers 1 ␮M each, and 1 U
Taq polymerase in the appropriate buffer. PCR condi-
tions were 30 to 35 cycles for 30 sec at 94°C, 30 sec at
the annealing temperature of the primers, and 30 sec
at 72°C. PCR products were separated on polyacryl-
amide gel in highly denaturing conditions (7 M urea,
32% formamide, and high voltage). After migration,
PCR products were transferred onto a nylon membrane
and fixed. Blotting and hybridization were performed
according to Hazan et al. [1992]. The poly(CA) probe
was labeled using 32P-dCTP or digoxigenine (Boeh-
 Genetic Counseling in Breast/Ovarian Cancer 177

TABLE I. Characteristics of the 23 Families Studied

Screened case
Family no. (age of onset) FBC(t)c FBC<45 BOvd OvC(t) MBC(t) U1Rt
101 FBC (36)a 3 (3) 3 0 0 0 2
102 FBC (30) 4 (3) 2 0 0 0 9
105 OvC (57) 1 (1) 0 0 2 (2) 0 4
201 OvC (40) 3 (1) 1 0 1 (1) 0 5
204 OvC (50) 2 (2) 2 1 1 (1) 0 7
206 OvC (45) 0 0 0 6 (2) 0 11
208 FBC (33) 7 (5) 4 0 0 0 9
209 MBC (39) 3 (2) 2 0 0 2 (2)e 6
301 FBC (46) 4 (4) 0 0 0 0 4
302 FBBC (30;47) 4 (3) 4 1 0 0 1
303 FBC (40) 3 (3) 2 0 0 0 4
304 OvC (45) 3 (3) 1 0 1 (1) 0 5
305 MBC (45) 4 (3) 2 0 1 2 (2) 2
306 FBC (36) 4 (3) 3 1 (1) 0 0 2
307 FBC (45) 4 (4) 0 0 0 0 2
311 OvC (46) 3 (3) 1 0 1 (1) 0 4
312 BBOv (33;36;50)+FBC (25)b 5 (3) 4 2 (1) 0 0 6
313 FBC (50) 6 (3) 3 0 0 0 3
404 FBC (39) 8 (2) 5 0 0 0 7
504 OvC (66) 4 (2) 4 0 1 (1) 0 2
506 OvC (43) 0 0 0 4 (3) 0 6
512 FBC (48) 4 (4) 0 0 0 0 5
520 OvC (49) 1 (1) 1 1 2 (1) 0 4
a
FBC, female breast cancer; OvC, ovarian cancer; MBC, male breast cancer; U1Rt, unaffected 1st-degree relatives typed
for markers; FBBC, female bilateral breast cancer; BBOv, bilateral breast and ovarian cancer.
b
Two screened cases (Stoppa-Lyonnet et al. [1996]).
c
(t), number of cases typed for markers; total number of FBC including BOv.
d
BOv, breast and ovarian cancer.
e
One bilateral male breast cancer.

ringer Mannheim). In one laboratory, one primer of
each marker was coupled with a fluorochrome and elec-
trophoresis was carried out on an ABI373A automate
(Applied Biosystems, Inc.).
BRCA1 and BRCA2 Mutation Detection
Three different methods were used for the BRCA1
screening of the coding exons and intron-exon junc-
tions: denaturing gradient gel electrophoresis (DGGE)
(families 101 to 105, and 301 to 313), a protein trunca-
tion test (PTT) on exon 11, and sequencing of the other
exons (families 204 to 210). PTT was performed accord-
ing to the conditions previously described by Hoger-
vorst et al. [1995], and DGGE as described by Stoppa-
Lyonnet et al. [1997]. Families 404 and 503 to 520 were
studied by completely sequencing the coding sequence
and the intron-exon junctions. BRCA1 mutations have
been previously reported in family 312 (identified as
family 355 [Stoppa-Lyonnet et al., 1996]), 302 (identi-
fied as family 15 [Stoppa-Lyonnet et al., 1997]), and in
family 520 (identified as family F417 [Sobol et al.,
1996]).
BRCA2 analysis was performed using a combination
of PTT and heteroduplex analysis (HDA). Exons 10 and
11 of BRCA2 were analyzed by PTT, and the rest of the
gene as well as the first 334 bp of exon 10 and the first
490 bp of exon 11 by HDA of lymphocyte genomic DNA.
The primer pairs used for PTT have been published
elsewhere [Serova et al., 1997]. The primers used to
amplify BRCA2 exons for heteroduplex analysis were
provided by R. Wooster. Heteroduplex analysis of PCR
fragments was performed using MDEª gel according to
the manufacturer’s instructions (Bioprobe Systems,
Bobigny, France). cDNA was available for four families
(301, 306, 307, and 311), and was screened by HDA
using primers published by Serova et al. [1997].
Only PCR products exhibiting an electrophoretic
variant pattern were directly sequenced on both
strands. BRCA1 variant DNA was sequenced with au-
tomated sequencing procedures (fluorometric method)
using Dye Terminator or Dye Primer Cycle Sequencing
Kit and an ABI373A automate (Perkin-Elmer Corp.).
BRCA2 variant DNA was sequenced with the USB
PCR product sequencing kit, according to the manufac-
turer’s instructions (Amersham).
The BRCA2 mutations have previously been re-
ported in families 305 and 313 (identified as families
Cu20 and Cu159 [Tatvigian et al., 1996]), and in fami-
lies 209 and 404 (identified as families CDF80 and
CRG32 [Serova-Sinilnikova et al., 1997]).
Risk Calculations
Posterior probability of linkage to BRCA1,2 us-
ing marker segregation. Multipoint LOD scores
Z(BRCA1) and Z(BRCA2) were calculated at the
BRCA1 and the BRCA2 locus respectively, using the
genetic model determined by Claus et al. [1991] and
slightly modified by Easton et al. [1993] for the BCLC
studies, and the LINKMAP program of the LINKAGE
package [Lathrop et al., 1984].
Assuming that a disease-causing allele segregates in
a family, posterior probabilities of linkage to BRCA1,
BRCA2, or BRCAx (without taking into account the
mutation screening results), ␻1, ␻2, and ␻x , respec-
178 Essioux et al.

tively, were calculated from the prior probabilities ␣1
and ␣2 [Ott, 1991]:
␣i10Zi
␻i =
␣110Z1 + ␣210Z2 + 共1 − ␣1 − ␣2兲
where i equals 1, 2, or x.
The proportion of families linked to BRCA1 or
BRCA2 estimated by the BCLC [Ford et al., 1998] were
used as prior probabilities. Five family groups were
considered. In each group the estimated proportion ␣1
and ␣2 of families attributable respectively to BRCA1
and BRCA2 were: 1) ␣1 ⳱ 0.32, ␣2 ⳱ 0.09 for families
with less than five cases of female breast cancer only; 2)
␣1 ⳱ 0.19, ␣2 ⳱ 0.66 for families with at least six cases
of female breast cancer; 3) ␣1 ⳱ 0.69, ␣2 ⳱ 0.19 for
families with one case of ovarian cancer; 4) ␣1 ⳱ 0.88,
␣2 ⳱ 0.12 for families with at least two cases of ovarian
cancer; 5) ␣1 ⳱ 0.19, ␣2 ⳱ 0.77 for families with at least
one case of male breast cancer.
Genetic risks. The probability R(BRCAi) (where i
equals 1 or 2) that an individual carries a mutation in
one of the two genes can be derived from the posterior
probability of linkage ␻i to one of the genes and the
probability R/BRCAi that the individual carries a
BRCAi disease-mutated allele given that this allele
segregates in the family. The formula is:
R(BRCAi) = (R/BRCAi)*␻i
R/BRCA1 and R/BRCA2 can be calculated using
the MLINK program of the LINKAGE package [Lath-
rop et al., 1984].
In families in which a mutation is found, it is possible
to evaluate the impact of heterogeneity on the ability of
the indirect approach to identify the carrier status of
relatives by comparing the risks R(BRCAi) and R/
BRCAi in first-degree relatives. Indeed, the latter prob-
ability is conditional on the locus involved in a given
family, and thus eliminates heterogeneity.
Probabilities of the family to be due to BRCA1,
BRCA2, or BRCAx given a negative mutation
screening result. The probabilities PN(BRCAi) that
a family is due to BRCAi given that a negative muta-
tion-screening result is obtained, were calculated using
the approach recently described by Serova et al. [1997].
This approach is a two-step procedure. First, the pos-
terior probability of linkage to BRCAi is calculated (see
above). Then, the probability that a mutation at a given
gene will not be found, even though we know that a
mutation of this gene actually segregates in the family,
is calculated by assuming a sensitivity of mutation de-
tection of 0.80 [Serova et al., 1996] for BRCA1 and
BRCA2 and by allowing the screened case to be spo-
radic. The probabilities PN(BRCAi) were calculated
with and without taking into account marker segrega-
tion to evaluate the significance of linkage information.
RESULTS AND DISCUSSION
Comparison of the Indirect Approach and
Mutation Screening
The results of the genetic risk calculations using
marker segregation are shown in Table II. Risks were
calculated in the screened case and in first-degree rela-
tives of the screened case to assess the efficiency of the
indirect approach as a symptomatic and a presymp-
tomatic genetic test. We classified the families accord-
ing to the probabilities R(BRCA1) or R(BRCA2) for the
screened case to be a mutation carrier: 1) risk 艌0.95,

TABLE II. Linkage-Based Probabilities of the Screened Cases and Their First-Degree Relatives to Be BRCA1 or BRCA2 Carriers
Compared With Mutation-Screening Results
Lod scores Screened case First-degree relative Screening result
Family no. Z(BRCA1) Z(BRCA2) R(BRCA1)a R(BRCA2)a R(BRCA1) R(BRCA2) BRCA1 BRCA2
101 0.10 0.21 0.34 0.17 0.33 0.16 Mutation
102 0.32 −0.27 0.49 0.04 0.02 0.05
105 −0.86 −0.40 0.67 0.31 0.21 0.16
201 0.48 −0.77 0.93 0.02 0.02 0.00 Mutation
204 0.50 0.10 0.89 0.06 0.89 0.00 Mutation
206 0.96 −0.21 0.99 0.01 0.92 0.00 Mutation
208b 1.57 — 0.88 — 0.88 — Mutation
209 0.17 −1.05 0.71 0.15 0.35 0.11 Mutation
301 −0.51 −0.13 0.10 0.10 0.05 0.02
302 0.53 −0.87 0.93 0.00 0.90 0.00 Mutation
303 −0.35 −0.14 0.05 0.07 0.04 0.05
304 0.63 0.69 0.73 0.24 0.68 0.03
305 −0.75 1.05 0.01 0.99 0.01 0.99 Mutation
306 0.24 0.24 0.72 0.20 0.66 0.00
307 0.36 −0.91 0.68 0.04 0.50 0.03
311 −1.23 −0.08 0.07 0.49 0.04 0.03
312 −0.37 −1.02 0.67 0.04 0.14 0.04 Mutation
313 0.51 0.97 0.08 0.87 0.04 0.82 Mutation
404 −0.50 0.44 0.03 0.88 0.01 0.87 Mutation
504 0.55 0.06 0.87 0.08 0.03 0.07 Mutation
506b 0.48 — 0.94 — 0.79 —
512 0.54 −0.51 0.58 0.02 0.51 0.01 Mutation
520 0.66 −0.60 0.99 0.01 0.57 0.00 Mutation
a
See text for definition.
b
Not typed for BRCA2 markers.
 Genetic Counseling in Breast/Ovarian Cancer 179

which means that the involvement of either BRCA1 or
BRCA2 is almost certain; 2) risk <0.95 but 艌0.80,
which means that this involvement is very likely; 3)
risk <0.80 and 艌0.20, which means that the situation
is ambiguous; 4) risks both <0.20, which means that
another gene is probably involved. Three families (206,
305, 520) belong to the first group, eight families (201,
204, 208, 302, 313, 404, 504, 506) to the second group,
ten families (101, 102, 105, 209, 304, 306, 307, 311, 312,
512) to the third group, and only two families (301, 303)
to the fourth group. This means that the screened case
is very likely a mutation carrier of BRCA1 in eight
families, of BRCA2 in three families, and of another
unknown gene in two families. A first-degree relative
had a risk of at least 0.95 in only one family (305, Fig.
1), and less than 0.95 but at least 0.80 in six families
(204, 206, 208, 302, 313, 404). Thus, the indirect ap-
proach was useful as a presymptomatic test in seven
families. Note that the carrier status of the unaffected
relative is more ambiguous than that of the screened
case, due to the different limitations of the indirect
approach, which will be further discussed.
The results of the mutation screening are in com-
plete agreement with the probabilities of the screened
case to be a mutation carrier of either BRCA1 or
BRCA2. Indeed, the expected mutation was found in
each of the three families belonging to the first group
(almost certain involvement of BRCA1 or BRCA2), and
in seven of the eight families belonging to the second
group (involvement very likely). A mutation was also
found in four of the ten families with ambiguous status
and in none of the two families in which another gene
was probably involved. This is in accordance with pre-
vious reports of mutation screening in families with
strong evidence of linkage [Serova et al., 1997]. The
mutations found in 14 of the 23 families (61%) are de-
scribed in Table III. Note that two different BRCA1
mutations were found in family 312 [Stoppa-Lyonnet et
al., 1996].
It is possible to evaluate the impact of heterogeneity
on the ability of the indirect approach to identify the
carrier status of relatives with these families. As
shown in Table III, the BRCA1/2 status of a first-
degree relative was almost certain (R/BRCA1 or R/
BRCA2 at least 0.95 or at most 0.05) in six families and
very likely (between 0.80 and 0.95 or 0.05 and 0.20) in
six families; i.e., six additional families other than the
six families mentioned above. This means that genetic
heterogeneity is partially the cause of the moderate
efficiency of the indirect approach as a genetic test.
Family 208 is a good illustration of the effect of genetic
heterogeneity (Fig. 2). The probability R(BRCA1) that
the screened case is a BRCA1 mutation carrier is only
88%, due to the low prior probability of linkage to
BRCA1 of breast cancer only families. Since five af-
fected women share the same BRCA1 haplotype, the
conditional probability R/BRCA1 reaches 0.99 in the
unaffected daughter.
The carrier status remains ambiguous in 14% of
families (2/14), showing that heterogeneity is not the
only factor responsible for the moderate efficiency of
the indirect approach. One other factor is the amount
of information provided by marker segregation in the
relatives. With two polymorphic markers flanking each
BRCA1/2 mutation, there were few limitations due to
marker homozygosity. The lack of information was
most often due to the limited number of available rela-
tives in a given family.
Another limitation of the indirect approach as a ge-
netic test is the rate of phenocopies among breast can-
cers. Several examples of sporadic cases, leading to
negative LOD scores with BRCA1 markers, were re-
ported in families with a BRCA1 mutation [Narod et
al., 1995; Friedman et al., 1995]. In our series, one

Fig. 1. Pedigree of family 305. Black circles and squares indicate affected individuals. Diagonal slashes indicate deceased. Numbers under the symbols
indicate age at death for deceased or age at last observation for living individuals (Br, breast cancer; Ov, ovarian cancer; Pro, prostate cancer; Bl, bladder
cancer; Per, peritoneum cancer). The numbers following the abbreviations indicate ages at diagnosis, where known. The numbers below the individual
symbols are marker alleles arranged into haplotypes. Marker alleles separated by a comma cannot be phased. The presence of a mutation is indicated
by an asterisk. The screened case is indicated by a plus sign.
180 Essioux et al.

TABLE III. BRCA1 and BRCA2 Mutations Identified

First-degree relative
Family no. BRCA1 mutation BRCA2 mutation R/BRCA1a R/BRCA2a
101 Ala1752Pro 0.94
201 1481insAG 0.02
204 1135insA 0.94
206 3958indelter 0.93
208 Arg1443ter 0.99
209 Gln2714terb 0.64
302 243delAc 0.96
d
305 8525delC 1.00
312 3600del11/4184del4e 0.14
313 9254del5d 0.92
404 Leu414terb 0.97
504 4184del4 0.04
512 Cys44Gly 0.81
520 5382insCf 0.57
a
Calculated among first-degree relatives of the screened case.
b
Mutations reported as CDF80 and CRG32 in Serova-Sinilnikova et al. [1997].
c
Mutation reported as 15 in Stoppa-Lyonnet et al. [1997].
d
Mutations reported as Cu20 and Cu159 in Tavtigian et al. [1996].
e
Mutations reported as 355 in Stoppa-Lyonnet et al. [1996].
f
Mutations reported as F417 in Sobol et al. [1996].

sporadic case probably exists in family 209 in which a
BRCA2 mutation segregates. This family has four
typed cases (two male breast cancers and two female
breast cancers). One of the affected women (breast can-
cer at age 40 years) does not share the BRCA2 haplo-
type common to the other three cases leading to a low
probability that BRCA2 is involved (Tables II and III).
Thus, in small families, the carrier risk computation is
sensitive to the occurrence of a phenocopy, although
this possibility is taken into account in the genetic
model. On the other hand, allowing for phenocopies in
the genetic model leads to ambiguity in the relation
between phenotypes and genotypes. Indeed, compared
to a genetic predisposition such as FAP, in which case
affected individuals are obligate mutation carriers, it
lowers the informativity of a family and thus limits the
use of the indirect approach.
Another example of misleading information provided
by segregation of markers is the existence of two dif-
ferent mutations in the same family (312) that could

Fig. 2. Pedigree of family 208. Symbols as in Figure 1 (Br, breast cancer; Col, colon cancer; Lu, lung cancer; Test, testis cancer). The N refers to the
absence of this mutation.
 Genetic Counseling in Breast/Ovarian Cancer 181

lead to dramatic errors in the inference on the carrier
status from segregation information [Stoppa-Lyonnet
et al., 1996], although this should be a rare event.
Significance Of Linkage Information in
Families With a Negative Mutation Result
The sample of nine families in which no mutation
was found consisted of four breast cancer only families,
four families with breast and ovarian cancers, and one
ovary-specific family. When a mutation is not found,
three conclusions are possible if one assumes that a
disease gene segregates in the family: 1) the detection
method lacks sensitivity, 2) the screened case is spo-
radic, and 3) another gene is involved (evidence for a
locus on 8p12-p22 has been reported by Kerangueven
et al. [1995] and by Seitz et al. [1997]).
Table IV shows the probabilities PN(BRCAi) that the
family is due to BRCAi given that no mutation was
found. These probabilities were calculated 1) without
the marker information (M−), and 2) with this infor-
mation (M+). The difference between these two prob-
abilities allows an evaluation of the information pro-
vided by marker segregation. This difference is at most
0.15 for all loci in five families (102, 301, 303, 306, 506),
reaches a value greater than 0.15 and at most 0.30 for
at least one locus in three families (105, 304, 307), and
reaches a value greater than 0.30 in only one family
(311). Thus, the difference is the lowest for the female
breast cancer only families. Indeed, these families are
small (no more than four cases) and this leads to inad-
equate linkage information.
We can also classify the families according to the
value of the risk, given all the available information,
that a mutation on BRCA1 or BRCA2 segregates, using
the same groups as above. A BRCA1 mutation is al-
most certainly involved in family 506 (PN(BRCA1 ⳱
0.97)). Indeed, since RNA was not available for screen-
ing in this family, a mutation outside the coding region
may have been missed. There are no families in the
second group (risk between 0.80 and 0.95), six families
(102, 105, 304, 306, 307, 311) in the ambiguous group
(risk between 0.20 and 0.80), and two families (301,
303) in the group in which another gene is probably
involved. It should be borne in mind that the probabili-
ties PN(BRCAi) were calculated assuming a causative
mutation in the families. However, it cannot be ex-
cluded that some families are not aggregations of spo-
radic cases.
Two parameters have to be discussed since they may
affect the risk values. The first one is the proportion of
families linked to BRCA1, BRCA2, or BRCAx. Indeed,
these proportions could affect the risk computation
since, in the absence of marker information, the poste-
rior probabilities of linkage are simply equal to the
prior ones. As mentioned above, these proportions were
estimated from the BCLC families. Since these families
were selected for linkage, the estimated proportions
might not be valid for our sample. This potential bias
was accounted for by stratifying according to the family
type. Another parameter is the sensitivity of mutation
detection. The value of 0.80 that was used for the risk
computation is the highest reported value, estimated
from a set of 20 highly informative families shown to be
linked to BRCA1 [Serova et al., 1996] and might not be
valid for the present sample. As a matter of fact, these
families could segregate mutations difficult to detect.
In this case, the negative screening result would bring
very little information and the risks would be those
computed in the first part of this section. The conclu-
sions would however be quite similar.
CONCLUSIONS
Based on a series of 23 families with at least three
cases of breast and/or ovarian cancer (which is a very
frequent situation that oncogeneticists have to deal
with), our results show that the information provided
by an indirect approach is quite variable among fami-
lies. Although some of the data used in this study have
already been published, our results allow us to quan-
tify the amount of information provided by the indirect
approach from a whole series. When no mutation is
found in a family, this approach appears reasonable
only if the family is large enough, with several affected
cases (more than three) and several close relatives
(more than three) available for DNA typing. Small
families may be an aggregation of sporadic cases and
are not informative enough for conclusions to be drawn.

TABLE IV. Probabilities of BRCA1, BRCA2, and BRCAx in Families in Which No

Mutation Is Found
PN(BRCA1) PN(BRCA2) PN(BRCAx)
a
Family no. M−a M+b M− M+b M−a M+b
Breast-ovary cancer families
105 0.86 0.68 0.14 0.32 0.00 0.00
304 0.47 0.65 0.14 0.22 0.39 0.13
306 0.48 0.57 0.14 0.17 0.26 0.38
311 0.50 0.06 0.15 0.24 0.35 0.70
Ovary-specific family
506 0.91 0.97 0.09 0.03 0.00 0.00
Breast cancer only families
102 0.10 0.20 0.04 0.02 0.86 0.78
301 0.11 0.04 0.04 0.03 0.85 0.93
303 0.19 0.10 0.08 0.06 0.74 0.84
307 0.10 0.36 0.04 0.03 0.86 0.61
a
Without taking into account linkage information.
b
Taking into account linkage information.
182 Essioux et al.
In practice, we propose that mutation screening is first
undertaken. If the result is negative and the family is
large enough, an indirect approach or a screening of
other relatives (to insure that the first screened case
was not a phenocopy), can be considered. In small fami-
lies, little help is to be expected from an indirect ap-
proach, which may be unnecessarily distressing.
ACKNOWLEDGMENTS
This work was supported by grants from Axe Géné-
tique et Cancer de la Ligue Nationale Contre le Cancer
(LNCC), the Comités Départementaux of Puy de Dôme,
Bouches-du Rhône, Alpes de Hautes-Provence, Var,
and Val de Marne of the LNCC, by the Association pour
la Recherche sur le Cancer (ARC), by the Programme
Hospitalier de Recherche Clinique (IDF94014), the
Fondation de France, the Caisse Nationale d’Assu-
rance Maladie, CAMPLP, GREG, the Association
Charbonnières Triathlon, FEGEFLUC, and from IN-
SERM 4U003C. L.E. is supported by a grant from the
ARC and the IFSBM (Institut Supérieur de Formation
aux Sciences BioMédicales). O.S. is supported by a
grant of the Comité de l’Ain of the LNCC. C.G. is sup-
ported by a grant from the LNCC. We thank all the
clinicians and molecular geneticists who participated
in the study: F. Allione, J.-M. Allouche, M. Auvray, P.
Berthet, D. Birnbaum, J.-Y. Bobin, P. Bougnoux, H.
Brandone, A. Bremond, B. Bressac-de-Paillerets, V.
Chamouton, D. Cohen, M. Conte, V. Couleau, J.-C.
Cuillière, A. Cussac, L. Dannel-Moor, D. Dargent, J.
Dauplat, B. Delafontan, B. Delpech, C. Digabel-
Chabay, S. Dixneuf, J.-C. Durand, I. Eugène, V. Feillet,
D. Fiere, A. Fourquet, D. Frappaz, M. Frenay, J.-P.
Fricker, P. Fumoleau, J.-P. Gerard, P. Gesta, J. Gio-
anni, Y. Guillard, A. Hardouin, J. Jacquemier, J.-P.
Jacquin, R. Lidereau, J.-M. Limacher, A. Lortholary,
M. Machelard-Roumagnac, H. Mignotte, M. Mousseau,
M. Musset, G. Netter-Pinon, M. Nezri, T. Noguchi, C.
Noguès, G. Perrot, P. Pouillart, V. Quillien, C. Racinet,
M. Resbeut, H. Robin, S. Saez, R. Salmon, R. Schaerer,
G. Thomas, N. Tournemaire, P. Validire, P. Vennin, P.
Vincent, and P. Zlattof. We are grateful to Nadine An-
drieu, Dominique Maraninchi, Claude Mawas, Thierry
Philip, and Richard Sauvan for helpful discussion and
to Lorna Saint-Ange for editing the manuscript. We are
indebted to the patients and their families for partici-
pating in this study.
REFERENCES
Claus EB, Risch N, Thompson WD (1991): Genetic analysis of breast cancer
in the cancer and steroid hormone study. Am J Hum Genet 48:232–
242.
Couch FJ, Weber BL, and the Breast Cancer Information Core (1996):
Mutations and polymorphisms in the familial early-onset breast cancer
(BRCA1) gene. Hum Mutat 8:8–18.
Easton DF, Bishop DT, Ford D, Crockford GP, and the Breast Cancer
Linkage Consortium (1993): Genetic linkage analysis in familial breast
and ovarian cancer: Results from 214 families. Am J Hum Genet 52:
678–701.
Ford D, Easton DF, Stratton M, Narod S, Goldgar D, Devilee P, Bishop DT,
Weber B, Lenoir G, Chang-Claude J, Sobol H, Teare MD, Struewing J,
Arason A, Scherneck S, Peto R, Rebbeck RR, Tonin P, Neuhausen S,
Barkardottir R, Eyfjord J, Lynch H, Ponder BAJ, Gayther SA, Birch
JM, Lindblom A, Stoppa-Lyonnet D, Bignon Y-J, Borg A, Hamman U,
Haites N, Scott RJ, Maugard CM, Vasen H, Seitz S, Cannon-Albright
LA, Schofield A, Hedman MA, and the Breast Cancer Linkage Consor-
tium (1998): Genetic heterogeneity and penetrance analysis of the
BRCA1 and BRCA2 genes in breast cancer families. Am J Hum Genet
62:676–689.
Friedman LS, Szabo CI, Ostermeyer EA, Dowd P, Butler L, Park T, Lee
MK, Goode EL, Rowell SE, King M-C (1995): Novel inherited mutations
and variable expressivity of BRCA1 alleles, including the founder mu-
tation 185delAG in Ashkenazi Jewish families. Am J Hum Genet 57:
1284–1297.
Hall JM, Lee MK, Newman B, Morrow JE, Anderson LA, Huey B, King
M-C (1990): Linkage of early-onset familial breast cancer to chromo-
some 17q21. Science 250:1684–1689.
Hazan J, Dubay C, Pankowiac MP, Becuwe N, Weissenbach J (1992): A
genetic linkage map of human chromosome 20 composed entirely of
microsatellite markers. Genomics 12:183–189.
Hogervorst FBL, Cornelis RS, Bout M, van Vliet M, Oosterwijk JC, Olmer
R, Bakker B, Klijn JGM, Vasen HFA, Meijers-Heijboer H, Menko FH,
Cornelisse CJ, den Dunnen JT, Devilee P, van Ommen G-JB (1995):
Rapid detection of BRCA1 mutations by the protein truncation test.
Nat Genet 10:208–212.
Kerangueven F, Essioux L, Dib A, Noguchi T, Allione F, Geneix J, Longy
M, Lidereau R, Eisinger F, Pébusque M-J, Jacquemier J, Bonaı̈ti-Pellié
C, Sobol H, Birnbaum D (1995): Loss of heterozygosity and linkage
analysis in breast carcinoma: Indication for a putative third suscepti-
bility gene on the short arm of chromosome 8. Oncogene 10:1023–1026.
Lathrop GM, Lalouel JM, Julier C, Ott J (1984): Strategies for multilocus
linkage analysis in humans. Proc Natl Acad Sci USA 81:3443–3446.
Mazoyer S, Dunning AM, Serova O, Dearden J, Puget N, Healey CS, Gay-
ther SA, Mangion J, Stratton MR, Lynch HT, Goldgar DE, Ponder BAJ,
Lenoir GM (1996): A polymorphic stop codon in BRCA2. Nat Genet
14:253–254.
Miki Y, Swensen J, Shattuck-Eidens D, Futreal PA, Harsham K, Tavtigian
S, Liu Q, Cochran C, Bennett LM, Ding W, Bell R, Rosenthal J, Hussey
C, Tran T, McClure M, Frye C, Hattier T, Phelps R, Haugen-Strano A,
Katcher H, Yakumo K, Gholami Z, Shaffer D, Stone S, Bayer S, Wray
C, Bogden R, Dayananth P, Ward J, Tonin P, Narod S, Bristow PK,
Norris FH, Helvering L, Morrison P, Rosteck P, Lai M, Barrett JC,
Lewis C, Neuhausen S, Cannon-Albright L, Goldgar D, Wiseman R,
Kamb A, Skolnick MH (1994): A strong candidate for the breast and
ovarian cancer susceptibility gene BRCA1. Science 266:66–71.
Narod S, Ford D, Devilee P, Barkardottir RB, Eyjford J, Lenoir G, Serova
O, Easton D, Goldgar D and the Breast Cancer Linkage Consortium
(1995): Genetic heterogeneity of breast-ovarian cancer revisited (1995):
Am J Hum Genet 57:957–958.
Ott J (1991): ‘‘Analysis of Human Genetic Linkage.’’ Baltimore: Johns Hop-
kins University Press, pp 250–255.
Puget N, Torchard D, Serova-Sinilnikova OM, Lynch HT, Feunteun J,
Lenoir GM, Mazoyer S (1997): A 1-kb alu-mediated germ-line deletion
removing BRCA1 exon 17. Cancer Res 57:828–831.
Rebbeck TR, Couch FJ, Kant J, Calzone K, DeShano M, Peng Y, Chen K,
Garber JE, Weber BL (1996): Genetic heterogeneity in hereditary
breast cancer: Role of BRCA1 and BRCA2. Am J Hum Genet 59:547–
553.
Rowell S, Newman B, Boyd J, King M-C (1994): Inherited predisposition to
breast and ovarian cancer. Am J Hum Genet 55:861–865.
Sambrook J, Fritsch EF, Maniatis T (1989): ‘‘Molecular Cloning: A Labo-
ratory Manual.’’ Cold Spring Harbor, NY: Cold Spring Harbor Labo-
ratory Press.
Seitz S, Rohde K, Bender E, Nothnagel A, Kölble K, Schlag PM, Scherneck
S (1997): Strong indication for a breast cancer susceptibility gene on
chromosome 8p12-p22 : Linkage analysis in German breast cancer
families. Oncogene 14:741–743.
Serova O, Montagna M, Torchard D, Narod SA, Tonin P, Sylla B, Lynch
HT, Feunteun J, Lenoir GM (1996): A high incidence of BRCA1 muta-
tions in 20 breast-ovarian cancer families. Am J Hum Genet 58:42–51.
Serova OM, Mazoyer S, Puget N, Dubois V, Tonin P, Shugart YY, Goldgar
D, Narod SA, Lynch HT, Lenoir GM (1997): Mutations in BRCA1 and
BRCA2 in breast cancer families: Are there more breast cancer-
susceptibility genes? Am J Hum Genet 60:486–495.
Serova-Sinilnikova OM, Boutrand L, Stoppa-Lyonnet D, Bressac-de-
Paillerets B, Dubois V, Lasset C, Janin N, Bignon YJ, Longy M, Mau-
gard C, Lidereau R, Leroux D, Frebourg T, Mazoyer S, Lenoir GM
(1997): BRCA2 mutations in hereditary breast and ovarian cancer in
France. Am J Hum Genet 60:1236–1239.
 Genetic Counseling in Breast/Ovarian Cancer
Sobol H, Stoppa-Lyonnet D, Bressac-de-Paillerets B, Peyrat J-P,
Kerangueven F, Janin N, Noguchi T, Eisinger F, Guinebretiere J-M,
Jacquemier J, Birnbaum D (1996): Truncation at conserved terminal
regions of BRCA1 protein is associated with highly proliferating he-
reditary breast cancers. Cancer Res 56:3216–3219.
Spirio L, Nelson L, Ward K, Burt R, White R, Leppert M (1993): A CA-
repeat polymorphism close to the adenomatous polyposis coli (APC)
gene offers improved diagnostic testing for familial APC. Am J Hum
Genet 52:286–296.
Stoppa-Lyonnet D, Fricker JP, Essioux L, Pagès S, Limacher JM, Sobol H,
Laurent-Puig P, Thomas G (1996): Segregation of two BRCA1 muta-
tions in a single family. Am J Hum Genet 59:479–481.
Stoppa-Lyonnet D, Laurent-Puig P, Essioux L, Pagès S, Ithier G, Ligot L,
Fourquet A, Salmon RJ, Clough KB, Pouillart P, the Institut Curie
Breast Cancer Group, Bonaı̈ti-Pellié C, Thomas G (1997): BRCA1 se-
quence variations in 160 individuals referred to a breast/ovarian family
cancer clinic. Am J Hum Genet 60:1021–1030.
Tavtigian SV, Simard J, Rommens J, Couch F, Shattuck-Eidens D, Neu-
hausen S, Merajver S, Tholarcius S, Offit K, Stoppa-Lyonnet D, Be-
langer C, Bell R, Berry S, Bogden R, Chen Q, Davis T, Dumont M, Frye
C, Hattier T, Jammulapati S, Janecki T, Jiang P, Kehrer R, Leblanc
J-F, Mitchell JT, McArthur-Morrison J, Nguyen K, Peng Y, Samson C,
Schroeder M, Snyder SC, Steele L, Stringfellow M, Stroup C, Swedlund
183
B, Swensen J, Teng D, Thomas A, Tran T, Tran T, Tranchant M,
Weaver-Feldhaus J, Wong AKC, Shizuya H, Eyfjord JE, Cannon-
Albright L, Labrie F, Skolnick MH, Weber B, Kamb A, Goldgar DE
(1996): The complete BRCA2 gene and mutations in chromosome 13q-
linked kindreds. Nat Genet 12:333–337.
Weissenbach J, Gyapay G, Dib C, Vignal A, Morissette J, Millasseau P,
Vaysseix G, Lathrop M (1992): A second-generation linkage map of the
human genome. Nature 359:794–801.
Wooster R, Neuhausen SL, Mangion J, Quirk Y, Ford D, Collins N, Nguyen
K, Seal S, Tran T, Averill D, Fields P, Marshall G, Narod S, Lenoir GM,
Lynch H, Feunteun J, Devilee P, Cornelisse CJ, Menko FH, Daly PA,
Ormiston W, McManus R, Pye C, Lewis CM, Cannon-Albright LA, Peto
J, Ponder BAJ, Skolnick MH, Easton DF, Goldgar DE, Stratton MR
(1994): Localization of a breast cancer susceptibility gene, BRCA2, to
chromosome 13q12-13. Science 265:2088–2090.
Wooster R, Bignell G, Lancaster J, Swift S, Seal S, Mangion J, Collins N,
Gregory S, Gumbs C, Micklen G, Bartfoot R, Hamoudi R, Patel S, Rice
C, Biggs P, Hashim Y, Smith A, Connor F, Arason A, Gudmundsson J,
Ficenec D, Kelsell D, Ford D, Tonin P, Bishop DT, Spurr NK, Ponder
BAJ, Eeles R, Peto J, Devilee P, Cornelisse C, Lynch H, Narod S, Lenoir
G, Egilsson V, Barkadottir RB, Easton DF, Bentley DR, Futreal PA,
Ashworth A, Stratton MR (1995): Identification of the breast cancer
gene BRCA2. Nature 378:789–792.