Context.—Several studies report the role of dimeric inhibin-A in assessing risk for fetal Down syndrome. The majority, however, use the Serotec inhibin-A assay and not the newer Diagnostic Systems Laboratories inhibin-A enzyme-linked immunosorbent assay (ELISA).

Objectives.—To establish normal gestational age day-specific medians, to compare our results against previous studies pertaining to the inhibin-A ELISA, and to evaluate long-term assay performance.

Design.—Using the inhibin-A ELISA, 100 specimens were assayed for each completed week of gestation for weeks 15 to 20, 50 specimens for 14 weeks, and 54 specimens for 21 weeks or older. Regressed inhibin-A medians were calculated employing a second-degree polynomial fit of the arithmetic medians. Thereafter, inhibin-A ELISA lot comparisons were performed to evaluate consistency.

Results.—Regressed values of 182, 174, 175, 184, 201, and 226 pg/mL resulted for weeks 15 to 20, respectively [pg/mL inhibin-A = 4.1528(gestational age)2 − 136.49(gestational age) + 1294.9]. A comparison with 2 other studies shows our values to be lower overall by 15 ± 11.4% and 16 ± 2.6%. However, variability between kit lots was as high as 30%.

Conclusions.—The equation derived provides for the calculation of gestational age day-specific inhibin-A medians for integration into maternal serum screening programs with a subsequent decrease in false-positives expected and observed. Our medians differ considerably from those of other studies, with limited data, using the Diagnostic Systems assay. However, lot changes since the initial analysis have exhibited similar inconsistencies. Therefore, we recommend that others incorporating the assay into their screening programs carefully establish, monitor, and adjust their medians accordingly as a result of potential variations.

Inhibin-A is a 32-kd glycoprotein hormone secreted by the ovaries, testis, and the placenta.1–3 The active form of the hormone, known as dimeric inhibin-A (DIA), consists of two protein subunits designated as α and βA and is characterized by its ability to inhibit the secretion of follicle-stimulating hormone.1 

In women, DIA is produced by the corpus luteum during the menstrual cycle and in early pregnancy.4 Although the function of DIA in pregnancy is unknown, it is suggested that DIA may be involved in the establishment of pregnancy, including fetal and placental development.5 

During pregnancy, the secretion of DIA by the placenta and other sources, such as the fetus, placental and fetal membranes, and the ovaries, augments circulating levels in the maternal serum.3–5 Maternal serum DIA levels rise during the first trimester and then decline after approximately 10 weeks. The maternal levels then remain fairly stable at 15 to 25 weeks of the pregnancy and then increase, reaching maximum levels at term.4 

Several studies have revealed a correlation between second-trimester maternal serum DIA concentration and fetal Down syndrome, with affected subjects having an approximately 2-fold or greater increase.4,6–10 Dimeric inhibin-A has also been shown to be of value in predicting other aneuploidies, such a trisomy 18.11–15 

Early inhibin-A assays had difficulty in identifying DIA from precursor forms of the protein, free circulating α-subunits, and inhibin-B, thus producing inconsistencies in screening performance.16,17 However, with the advent of a monoclonal antibody pair specific for DIA, the ability to exclusively measure DIA is now feasible.18–21 

Two commercial assays specific for DIA have generally been used: the Serotec Inhibin-A Dimer Assay Kit (Serotec Ltd, Kidlington, Oxford, United Kingdom) and the Diagnostic Systems Laboratories (DSL) ACTIVE Inhibin-A enzyme-linked immunosorbent assay (ELISA) (Diagnostic Systems Laboratories Inc, Webster, Tex). The advantages of using the DSL assay are that serum pretreatment by boiling in sodium dodecyl sulfate is not required and pretreatment with hydrogen peroxide is performed directly in the microwell of the 96-well microtiter plate, thus significantly reducing assay time and labor.22 

The majority of published studies have used the Serotec assay and not the DSL assay. In addition, the few studies using the DSL assay lack adequate observations at various weeks of gestational age (GA), with as few as 4 observations for a given week (Table 1). Therefore, the purpose of our study was 3-fold. First, we wanted to establish our own normal day-specific GA-based medians for the DSL DIA assay for incorporation into a maternal serum quadruple screening test (α-fetoprotein, human chorionic gonadotropin, unconjugated estriol, and DIA) for fetal Down syndrome. Second, we sought to compare and evaluate our resultant medians with those of 2 recently published studies that also used the DSL DIA assay.22,23 Third, we wanted to evaluate the lot-to-lot consistency of the DSL assay in order to implement contingencies for maintaining reliability in reporting fetal Down syndrome risk. All of these objectives were achieved.

Table 1.

Regressed Dimeric Inhibin-A (DIA) Median Comparison

Regressed Dimeric Inhibin-A (DIA) Median Comparison
Regressed Dimeric Inhibin-A (DIA) Median Comparison

ACTIVE Inhibin-A ELISA kits were purchased from Diagnostic Systems Laboratories Inc. The SPECTRAmax PLUS plate reader was manufactured by Molecular Devices Corporation (Sunnyvale, Calif) and PC controlled using Molecular Devices ProMax software. Data analysis was performed using Microsoft Corporation Excel software (Bellevue, Wash).

Second-trimester maternal serum specimens, deemed to be normal from triple test screening results (α-fetoprotein, human chorionic gonadotropin, and unconjugated estriol), were collected and stored at −70°C. Specimens were restricted to singleton, nondiabetic, non-black pregnancies because of the known variations caused by multiple pregnancies, maternal diabetes, and race.24–27 Specimen criteria were also based on weeks of GA completed, with a minimum of 100 specimens each for weeks 15 to 20 and a minimum of 50 specimens for 14 weeks and 21 weeks or older. Specimens were de-identified under Internal Review Board protocol 7275.

Specimens were thawed and then analyzed in duplicate for DIA using the DSL ELISA assay according to the manufacturer's instructions. (The assay was previously validated and offered in our laboratory for monitoring reproductive physiology.) Assay imprecision was monitored using the manufacturer's level I and level II controls and a previously established in-house high control consisting of pooled serum.

Dimeric inhibin-A assay results were categorized according to completed weeks of GA and the arithmetic median calculated for each completed week. A plot of the arithmetic median DIA concentration versus GA week completed was constructed for weeks 15 to 20, and the regressed DIA values were determined using a second-degree polynomial fit of the data. The fit was restricted to 15 to 20 weeks of GA to include only evenly weighted data and to achieve a direct week-to-week comparison with 2 other studies listing GA-specific medians for the DSL assay.22,23 

Although practically all maternal serum screening specimens shipped to our laboratory are received and stored frozen, specimens are subjected to extended periods of refrigeration after thawing because of the number of tests executed for patient risk assessment. Dimeric inhibin-A stability at 4°C was investigated by aliquoting 3 patient serum specimens of low, mid-, and high DIA concentrations and allowing designated aliquots to sit at 4°C for intervals of 0, 1, 2, 4, and 7 days and then freezing. The aliquots were later thawed and measured simultaneously for DIA using the DSL ELISA.

After implementing our maternal serum quad screening program (α-fetoprotein, human chorionic gonadotropin, unconjugated estriol, and DIA), DSL DIA ELISA lot-to-lot variation was investigated. Subsequent kit lots were compared by selecting a wide DIA concentration range of specimens previously measured with the older lot and reassaying using the newer lot. A total of 7 lots were evaluated by measuring 10 to 34 specimens per lot change, depending on how well the data correlated. Data analysis was performed by Deming regression, plotting the new lot results versus the old. The medians for the new lot were then adjusted accordingly for each week of GA using the resultant fit equation from the Deming plot. The difference in median values from the previous lot to the next was calculated for each week of GA and expressed as a percentage.

Measured DIA concentrations ranged from 50 to 887 pg/mL, with a mean of 215 pg/mL DIA (n = 704). Assay precision, as monitored over 19 runs using the 3 controls, was as follows: DSL level I, 55 ± 5; DSL level II, 276 ± 21; and in-house pooled high, 648 ± 42 pg/mL DIA, generating coefficients of variation of 8.9%, 7.6%, and 6.4%, respectively (Figure 1). The three 4°C stability specimens generated values of 62 ± 2, 424 ± 14, and 839 ± 20 pg/mL DIA for the low-, mid-, and high-level specimens, respectively. All measurements fell within 95% confidence (2 SD) of their respective means, as shown in Figure 2.

Figure 1.

Diagnostic Systems Laboratories (DSL) dimeric inhibin-A (DIA) enzyme-linked immunosorbent assay precision. Assay precision was monitored using the 2 DSL kit controls and an in-house control comprising pooled serum. The solid and dashed lines represent the means and 2 SD, respectively. Results: level I, 55 ± 5; level II, 276 ± 21; and pooled high, 648 ± 42 pg/mL DIA, with coefficients of variation of 8.9%, 7.6%, and 6.4%, respectively.

Figure 1.

Diagnostic Systems Laboratories (DSL) dimeric inhibin-A (DIA) enzyme-linked immunosorbent assay precision. Assay precision was monitored using the 2 DSL kit controls and an in-house control comprising pooled serum. The solid and dashed lines represent the means and 2 SD, respectively. Results: level I, 55 ± 5; level II, 276 ± 21; and pooled high, 648 ± 42 pg/mL DIA, with coefficients of variation of 8.9%, 7.6%, and 6.4%, respectively.

Close modal
Figure 2.

Dimeric inhibin-A (DIA) stability at 4°C. Three maternal serum specimens were aliquoted and refrigerated at 4°C. One aliquot from each level was then frozen after 0, 1, 2, 4, and 7 days of storage. All samples were simultaneously thawed and assayed for DIA. The solid and dashed lines represent the means and 2 SD, respectively. Excellent DIA stability is exhibited up to the 7-day test period

Figure 2.

Dimeric inhibin-A (DIA) stability at 4°C. Three maternal serum specimens were aliquoted and refrigerated at 4°C. One aliquot from each level was then frozen after 0, 1, 2, 4, and 7 days of storage. All samples were simultaneously thawed and assayed for DIA. The solid and dashed lines represent the means and 2 SD, respectively. Excellent DIA stability is exhibited up to the 7-day test period

Close modal

The arithmetic DIA medians were 229, 174, 191, 170, 177, 201, 229, 229, 197, 243, and 379 pg/mL DIA for completed weeks of GA 14 through 24, respectively. A second-degree polynomial fit of the arithmetic medians for completed weeks of GA 15 to 20 generated equation 1 below and is illustrated in Figure 3:

pg/mL DIA = 4.1528 × GA2 − 136.49 × GA >+ 1294.9 (R2 = 0.8194)
(1)
Figure 3.

Regressed dimeric inhibin-A (DIA) medians. DIA was measured for 100 normal maternal serum specimens for each completed week, gestational age (GA) 15 through 20. A second-degree polynomial fit of the DIA concentration versus GA generated the equation. Day-specific normal DIA medians are calculated by inserting GA values to the nearest 0.01 week

Figure 3.

Regressed dimeric inhibin-A (DIA) medians. DIA was measured for 100 normal maternal serum specimens for each completed week, gestational age (GA) 15 through 20. A second-degree polynomial fit of the DIA concentration versus GA generated the equation. Day-specific normal DIA medians are calculated by inserting GA values to the nearest 0.01 week

Close modal

The regressed DIA median values calculated from equation 1 are as follows: 182, 174, 175, 184, 201, and 226 pg/mL DIA for 15 to 20 weeks of GA, respectively (Table 1). A week-to-week comparison of this study's regressed medians with those of Knight et al22 and MacRae et al,23 as listed in Table 1, shows our values to be lower overall by 15 ± 11.4% and 16 ± 2.6%, respectively. It is noted that the Knight et al22 medians were calculated using a corrected equation, developed following communication with the authors, and not with the equation listed in their article.

The Deming regression equations of the new versus the old generated for each lot change are listed in Table 2. The slopes vary from 0.79 to 1.20 and intercepts from −18.19 to 53.52. The correlation coefficients were 0.95 or greater. After adjustment of the median values for each week of GA, median differences from the previous lot to the next varied from −12% to 30%, as demonstrated in Figure 4.

Table 2.

Dimeric Inhibin-A Lot-to-Lot Deming Regression Results

Dimeric Inhibin-A Lot-to-Lot Deming Regression Results
Dimeric Inhibin-A Lot-to-Lot Deming Regression Results
Figure 4.

Lot-to-lot variability of the Diagnostic Systems Laboratories (DSL) dimeric inhibin-A (DIA) enzyme-linked immunosorbent assay (ELISA). Maternal serum specimens were assayed for DIA using both the old and new kit lots. A plot was constructed of the new versus the old lot and was analyzed by Deming regression. Medians were then adjusted according to the fit equation generated. For simplicity, only the 17th week median between each lot is represented

Figure 4.

Lot-to-lot variability of the Diagnostic Systems Laboratories (DSL) dimeric inhibin-A (DIA) enzyme-linked immunosorbent assay (ELISA). Maternal serum specimens were assayed for DIA using both the old and new kit lots. A plot was constructed of the new versus the old lot and was analyzed by Deming regression. Medians were then adjusted according to the fit equation generated. For simplicity, only the 17th week median between each lot is represented

Close modal

The equation derived in this study does provide a means for calculating GA day-specific (to the nearest 0.01 week of GA) DIA medians for utilization in a quad maternal screening test for assessing risk of Down syndrome. With the exception of the lot-to-lot variability, the DSL DIA ELISA did perform well throughout this study, as demonstrated by the precision shown in Figure 1, which was generated using the same kit lot, and by the fact that only 3 of the study's 704 specimens needed to be repeated. In addition, DIA was stable in serum no less than 7 days when stored at 4°C (Figure 2).

The regressed DIA medians established in this study, together with frequent monitoring and adjustment, have worked well in our maternal serum screening program. By incorporating DIA with the triple test markers, thus establishing a quad test, the initial positive rate for Down syndrome has decreased from 6% to around 2%, while maintaining the triple test detection rate of approximately 70%. Although an increase in the Down syndrome detection rate is achievable by the incorporation of DIA into maternal screening,4,9,10 we have chosen instead to lower the initial positive rate. This has led to an approximate 50% decrease in unnecessary ultrasound examinations and amniocentesis, with a corresponding reduction in the number of patients experiencing increased distress because of their screening outcomes.

As mentioned previously, we limited our results to 15 to 20 weeks of GA in order to achieve a direct week-to-week comparison with 2 other studies. However, our additional results of 229 pg/mL for week 14 and 229, 197, 243, and 379 pg/mL for weeks 21 through 24, respectively, do coincide with the trend observed previously—a trend in which DIA levels decline, stabilize, and then increase to a maximum at term.4 This adds reassurance that our results for the critical 15 to 20 weeks of GA coincide with previously observed trends and are suitable for comparison.

Table 1 shows that the 2 other studies addressing the DSL DIA assay do lack in terms of the number of observations for various weeks of GA, in particular, weeks 18 through 20 of the Knight study.22 Our study, with 100 specimens for each of the weeks 15 through 20, provides a greater trust in our resultant medians. Furthermore, the Knight study's day-specific median equation22 does contain an error, as confirmed by communication with the authors. The equation should stand alone and not as the exponent for the inverse log, as indicated in their article. This correction was made when calculating the regressed medians included in Table 1. The MacRae study23 does not provide an equation, but the article does list their regressed values, which are included in Table 1.

Comparison of the regressed medians of this study shows our initial medians overall to be 15 ± 11.4% and 16 ± 2.6% lower than those of Knight et al22 and MacRae et al,23 respectively. Moreover, the difference at specific weeks of GA (weeks 15 and 16) was as high as 25% for the Knight study.22 Based on this observation, we considered the possibility that changes in the ELISA kit manufacturing process during the time period between the studies above and our own could be a factor contributing to this observation. However, communication with DSL revealed that they had increased only their DIA ELISA kit manufacturing capacity during this time period, leaving the procedures themselves unchanged.

Since our initial study and the implementation of our quad screening program more than a year ago, we have experienced 7 DIA ELISA lot changes. As Table 2 and Figure 4 indicate, the disparity observed between regressed medians is most likely caused by the recurring inconsistencies between DSL DIA kit lots. In other words, with differences ranging from −12% to 30% between lots, the lot-to-lot variability almost certainly contributes more to our disagreement with the MacRae et al23 and Knight et al22 studies than do the limitations of those studies, as expressed previously. For this reason, we have had to, and continue to, diligently and painstakingly monitor and adjust our medians with each lot change, as dictated by Deming regression analysis.

In conclusion, the implementation of DIA in our maternal serum screening program does provide for a decrease in the initial positive rate for assessing risk of Down syndrome. Additionally, we have outlined a procedure for establishing DIA medians for use in quad screening programs, including the means of deriving an equation for calculating GA day-specific median values. Although the performance characteristics within individual DIA ELISA assay lots do function adequately, severe inconsistencies have been recognized among previous studies and continue to be observed between assay kit lots. Although a version of the Serotec DIA assay was offered at the time of our study, Serotec discontinued the assay at the end of 2003. This is a result of the purchase of their supplier, Oxford Bio-Inovation Ltd (Oxfordshire, United Kingdom), by DSL (information gleaned via communication with Technical Sales Support at Serotec Inc). Consequently, it is our understanding that the DSL ACTIVE Inhibin-A ELISA is the only commercial test available for measuring DIA at this time. Therefore, it is imperative that laboratories integrating the assay into their maternal screening programs meticulously establish, monitor, and modify their medians accordingly.

Operating and reagent costs were provided by the ARUP Institute for Clinical and Experimental Pathology, LLC, in Salt Lake City, Utah. Support was also received from the Institute's Research and Development, Special Chemistry section.

Ying
,
S. Y.
Inhibins, activins and follistatins: gonadal proteins modulating the secretion of follicle-stimulating hormone.
Endocrine Rev
1988
.
9
:
267
293
.
McLachlan
,
R. I.
,
D. L.
Healy
,
P. J.
Lutjen
,
J. K.
Findlay
,
D. M.
de Krester
, and
H. G.
Burger
.
The maternal ovary is not the source of circulating inhibin levels during human pregnancy.
Clin Endocrinol
1987
.
27
:
663
668
.
Muttukrishna
,
S.
,
T. J.
Child
,
N. P.
Groome
, and
W. L.
Ledger
.
Source of circulating levels of inhibin A, pro alpha C-containing inhibins and activin A in early pregnancy.
Hum Reprod
1997
.
12
:
1089
1093
.
Wald
,
N. J.
,
J. W.
Densem
,
L.
George
,
S.
Muttukrishna
, and
P. G.
Knight
.
Prenatal screening for Down's syndrome using inhibin-A as a serum marker.
Prenat Diagn
1996
.
16
:
143
153
.
Riley
,
S. C.
,
N. C.
Wathen
,
T.
Chard
,
N. P.
Groome
, and
E. M.
Wallace
.
Inhibin in extra-embryonic coelomic and amniotic fluids and maternal serum in early pregnancy.
Hum Reprod
1996
.
11
:
2772
2776
.
Spencer
,
K.
,
E. M.
Wallace
, and
S.
Ritoe
.
Second-trimester dimeric inhibin-A in Down's syndrome screening.
Prenat Diagn
1996
.
16
:
1101
10
.
Aitken
,
D. A.
,
E. M.
Wallace
, and
J. A.
Crossley
.
et al
.
Dimeric inhibin A as a marker for Down's syndrome in early pregnancy.
N Engl J Med
1996
.
334
:
1231
1236
.
Wenstrom
,
K. D.
,
J.
Owen
,
D. C.
Chu
, and
L.
Boots
.
Elevated second-trimester dimeric inhibin A levels identify Down syndrome pregnancies.
Am J Obstet Gynecol
1997
.
177
:
992
996
.
Haddow
,
J. E.
,
G. E.
Palomaki
,
G. J.
Knight
,
D. L.
Foster
, and
L. M.
Neveux
.
Second trimester screening for Down's syndrome using maternal serum dimeric inhibin A.
J Med Screen
1998
.
5
:
115
119
.
Benn
,
P. A.
,
M.
Fang
,
J. F.
Egan
,
D.
Horne
, and
R.
Collins
.
Incorporation of inhibin-A in second-trimester screening for Down syndrome.
Obstet Gynecol
2003
.
101
:
451
454
.
Wenstrom
,
K. D.
,
D. C.
Chu
,
J.
Owen
, and
L.
Boots
.
Maternal serum α-fetoprotein and dimeric inhibin A detect aneuploidies other than Down syndrome.
Am J Obstet Gynecol
1998
.
179
:
966
970
.
Lambert-Messerlian
,
G. M.
,
D. N.
Saller
,
M. B.
Tumber
,
C. A.
French
,
C. J.
Peterson
, and
J. A.
Canick
.
Second-trimester maternal serum inhibin A levels in fetal trisomy 18 and Turner syndrome with and without hydrops.
Prenat Diagn
1998
.
18
:
1061
1067
.
Wenstrom
,
K. D.
,
J.
Owen
,
D. C.
Chu
, and
L.
Boots
.
Prospective evaluation of free β-subunit of human chorionic gonadotropin and dimeric inhibin A for aneuploidy detection.
Am J Obstet Gynecol
1999
.
181
:
887
892
.
Benn
,
P. A.
,
A.
Gainey
,
C. J.
Ingardia
,
J. F.
Rodis
, and
J. F.
Egan
.
Second trimester maternal serum analytes in triploid pregnancies: correlation with phenotype and sex chromosome complement.
Prenat Diagn
2001
.
21
:
680
686
.
Watanabe
,
H.
,
H.
Hamada
, and
N.
Yamada
.
et al
.
Second-trimester maternal pregnancy-associated plasma protein a and inhibin A levels in fetal trisomies.
Fetal Diagn Ther
2002
.
17
:
137
141
.
Spencer
,
K.
,
P. J.
Wood
, and
F. W.
Anthony
.
Elevated levels of maternal serum inhibin immunoreactivity in second trimester pregnancies affected by Down's syndrome.
Ann Clin Biochem
1993
.
30
:
219
220
.
Chuckle
,
H. S.
,
S.
Holding
, and
R.
Jones
.
Maternal serum inhibin levels in second trimester Down's syndrome pregnancies.
Prenat Diagn
1994
.
14
:
387
390
.
Groome
,
N. P.
,
J.
Hancock
,
A.
Betteridge
,
M.
Lawrence
, and
R.
Craven
.
Monoclonal and polyclonal antibodies reactive with the 1–32 amino-terminal peptide of the alpha subunit of human inhibin.
Hybridoma
1990
.
9
:
31
42
.
Groome
,
N. P.
and
M.
Lawrence
.
Preparation of monoclonal antibodies reactive with the beta A subunit of human ovarian inhibin.
Hybridoma
1991
.
109
:
309
316
.
Groome
,
N. P.
and
M.
O'Brian
.
Immunoassays for inhibin and its subunits: further applications of the synthetic peptide approach.
J Immunol Methods
1993
.
165
:
167
176
.
Groome
,
N. P.
,
P. J.
Illingworth
, and
M.
O'Brian
.
et al
.
Detection of dimeric inhibin throughout the human menstrual cycle by two-site enzyme immunoassay.
Clin Endocrinol
1994
.
40
:
717
723
.
Knight
,
G. J.
,
G. E.
Palomaki
,
L. M.
Neveux
,
J. E.
Haddow
, and
G. M.
Lambert-Messerlian
.
Clinical validation of a new dimeric inhibin A assay suitable for second trimester Down's syndrome screening.
J Med Screen
2001
.
8
:
2
7
.
MacRae
,
A. R.
,
L. C.
Allen
,
N.
Lepage
, and
S.
Tokmakejian
.
Validation of an automated dimeric inhibin-A assay in maternal serum screening for Down syndrome.
Clin Chem
2001
.
47
:
A150
.
Watt
,
H. C.
,
N. J.
Wald
, and
L.
George
.
Maternal serum inhibin-A levels in twin pregnancies: implications for screening for Down's syndrome.
Prenat Diagn
1996
.
16
:
927
929
.
Watt
,
H. C.
,
N. J.
Wald
, and
L.
George
.
Maternal serum inhibin-A in pregnancies with insulin-dependent diabetes mellitus: implications for screening for Down's syndrome.
Prenat Diagn
1996
.
16
:
923
926
.
Wallace
,
E. M.
,
J. A.
Crossley
,
C. R.
Savitri
,
N. P.
Groome
, and
D. A.
Aitken
.
Maternal serum inhibin-A in pregnancies complicated by insulin dependent diabetes mellitus.
Br J Obstet Gynecol
1997
.
104
:
946
948
.
Watt
,
H. C.
,
N. J.
Wald
,
D.
Smith
,
A.
Kennard
, and
J.
Densem
.
Effect of allowing for ethnic group in prenatal screening for Down's syndrome.
Prenat Diagn
1996
.
16
:
691
698
.

The authors have no relevant financial interest in the products or companies described in this article.

Author notes

Reprints: J. Alan Erickson, PhD, ARUP Laboratories, 500 Chipeta Way, Salt Lake City, UT 84108 ([email protected])