The application of mass spectrometry has evolved considerably since its first use and mass spectrometric methods were initially introduced in laboratory medicine approximately 40 years ago [1]. The very recent popularity of clinical mass spectrometry can be attributed to the high specificity, accuracy and reliability due to the direct analysis of ions without the risk of cross reactivity as described for antibody detection in immunoassays [2] as well as the ability to detect multi-analytes in a single run. Initially, GC-MS was used for biological analysis, however, this method requires volatile analytes, demanding extensive extraction and derivatization steps for nonvolatile and thermally unstable compounds typically found in clinical analysis. This is not particularly attractive in a clinical setting, in contrast to LC-MS/MS which offers the advantages of mass spectrometry analysis in combination with a simpler sample preparation technique.

by Dr Nihâl Yüksekdag, Dr Marc Egelhofer and Dr Richard Lukacin

One such example is the analysis of methylmalonic acid (MMA), an important biomarker for the identification of vitamin B12 deficiency which, if left untreated, can lead in the long term to permanent neurological damage and/or to hematological and gastroenterological diseases. The sole determination of holoTC, the active form of vitamin B12,  does not have the same diagnostic significance as the combined measurement of holoTC and MMA, as the MMA concentration shows a possible vitamin B12 deficiency even before the actual vitamin level decreases [3]. Traditionally, the reference method for this parameter in plasma/serum is GC-MS which, as mentioned above, requires an extremely complex sample preparation that can take several hours [4]. In contrast to this, the sample preparation for LC-MS/MS from Chromsystems is much easier, and, with just a few minutes processing time, considerably faster, while requiring only one quarter of the sample material (see table 1).

Furthermore, data from plasma and urine MMA determinations by the reference GC-MS method and the new LC-MS/MS technology show a strong correlation and excellent agreement (Fig. 1). Therefore, the described LC-MS/MS technique represents a fast, reliable and robust method for  routine analysis, achieving a higher throughput and higher efficiency.

Sample preparation as a pivotal step
The correct analytical procedure from extraction and sample preparation, through to the chromatography and MS setup is a prerequisite to achieve optimal results by mass spectrometry, and to fulfil the requirements in clinical diagnostics. The development of an appropriate sample preparation procedure can be complicated and time-consuming, requiring considerable work in order to sensibly embed it in the overall analytical procedure. The ultimate goal is the enrichment of the molecule of interest by a simultaneous elimination of compounds that cause ion suppression or enhancement effects. Moreover, components from plastic, chemicals like salts or particularly from the human matrix (whole blood, serum, plasma, urine), potentially co-eluting from the LC system can compete with the analytes during the ionization process. This leads to a change in compound ionization, and consequently alters the MS signal at the detector [5]. This process is called “ion suppression” and Bonfiglio et al [6] systematically analysed these effects and have found not surprisingly that they are dependent on the sample preparation technique used as well as the compound to be analysed. More polar analytes also showed stronger effects than less polar ones. Short-term variations in ionization can also compromise the accuracy of analyses, if the method is not sufficiently robust. If these variations have a differential impact on the target analyte and internal standard, the overall analysis is affected [7]. The authors also concluded the need for calibration material to be as similar as possible to the sample matrix. In addition, the choice of an appropriate internal standard helps to reduce matrix effects; whenever possible, an isotopically labelled version of the analyte is the ideal choice.

Depending on sample specimens and analyte characteristics, sample preparation techniques can encompass liquid-liquid extraction, solid phase extraction or protein precipitation and are also crucial for the removal of materials that may contaminate the column, trap-column or the analytical system. 

Considering all of these factors, successful method development where all parameters work well within at least acceptable levels of CVs, recovery and appropriate limits of quantification (LOQs) can be very challenging.  Furthermore, full establishment of a method that is comprehensively validated in the laboratory is a laborious process. The use of commercially available kits, like the one mentioned above for MMA, which have gone through numerous optimization, verification and validation processes from sample preparation through to MS analysis represents a secure, robust and time-saving alternative for clinical laboratories.

Multi-analyte determination
The capability of LC-MS/MS systems for the analysis of several compounds in a single run sounds efficient and relevant, e.g. for the simultaneous analysis of drugs and their metabolites, but may not be as easy as it seems. Every single analyte in a patient sample may possess different chemical and physical properties that affect its recovery in the sample preparation procedure. Consequently, some compounds may be extracted more efficiently than others. Therefore, it can be a highly complex task with a significant amount of work to develop a general sample preparation procedure for quantification of numerous drugs and metabolites, with many of them being analysed in a single run (see Fig. 2), aimed at simplifying the laboratory workflow.

Automation for a higher throughput
One of the major challenges clinical laboratories have been facing is the simplification and acceleration of sample preparation for LC-MS/MS. By using an automated workflow potential pipetting errors can be minimized and, in parallel, the throughput can be drastically increased. This is relevant, for example, to large transplant centres that analyse a high number of patient samples for immunosuppressive drugs, but nevertheless need to achieve fast and reliable results by LC-MS/MS. To date, there is only one system on the market (MassSTAR) that allows a fully automated CE-IVD workflow for immunosuppressants including sample tracking, LIMS connectivity and clotting detection. The automated method offers a time saving  of approximately 80% compared to manual preparation. A comparison between manual and automated sample preparation and measurement techniques for the four immunosuppressants cyclosporine A, everolimus, sirolimus and tacrolimus showed very high correlations (Fig. 3). Automated and manual preparation procedures therefore produce almost the same results, with automation reducing the time needed for sample extraction while also increasing sample throughput. These automation options are also provided by Chromsystems for other parameters, such as vitamin D3/D2, the immunosuppressant mycophenolic acid and antiepileptics, for which comparable correlations between the manual and the automated methods have also been shown.

A gold standard in routine
LC-MS/MS is a valuable technique that is often used in reference methods for a wide range of parameters. Its main drivers for growth in clinical laboratories are the limitations of immunoassays for low molecular weight compounds, the easier workflows and higher throughput [8]. However, there are certain downfalls that need to be addressed with one of the most, or even the most critical factor in clinical mass spectrometry being the application of an appropriate sample preparation procedure that is robust as well as reliably fulfilling analytical requirements. A number of proven and CE-IVD approved LC-MS/MS kits for sample preparation from Chromsystems are available and simplify the workflow in the laboratory. Furthermore, automation is also possible for a range of parameters, reducing hands-on time and increasing throughput for those laboratories with the need for higher throughput.

References
1. Vogeser M, Kirchhoff F (2011) Progress in automation of LC-MS in laboratory medicine. Clin Biochem 44(1): 4-13.
2. Korecka M, Shaw L, (2009) Review of the newest HPLC methods with mass spectrometry detection for determination of immunosuppressive drugs in clinical practice. Ann. Transplant 14(2): 61-72.
3. Obeid R, (2014) Methylmalonic acid – a biomarker for vitamin B12 deficiency. DIALOG 1/2014.
4. Obeid R, Geisel J, Kirchhoff F, Bernhardt K, Ranke D, Lukačin R. (2014) External validation of a novel commercially available mass spectrometry kit for MMA in serum/plasma and urine. Poster presented at the congress of the International Federation of Clinical Chemistry and Laboratory Medicine (IFCC) WorldLab, Istanbul, Turkey.
5. Schneider H, Steimer W. (2006) Tandem mass spectrometry in drug monitoring: experience and pitfalls in application. J Lab Med 30(6): 428-437.
6. Bonfiglio R, King RC, Olah TV, Merkle K. (1999) The effects of sample preparation methods on the variability of the electrospray ionization response for model drug compounds. Rapid Commun Mass Spectrom 13(12): 1175-1185.
7. Vogeser M, Seger C. (2010) Pitfalls associated with the use of liquid chromatography-tandem mass spectrometry in the clinical laboratory. Clin Chem 56(8): 1234-1244.
8. Grebe S, Singh R. (2011) LC-MS/MS in the Clinical Laboratory – Where to From Here? Clin Biochem Rev 32: 5-31.

The authors
Nihâl Yüksekdağ PhD, Marc Egelhofer PhD*, and Richard Lukačin PhD.
Chromsystems Instruments & Chemicals GmbH, Am Haag 12, 82166 Gräfelfing,  Germany
*Corresponding author, egelhofer@chromsystems.de

The first recorded outbreaks of Ebola Virus Disease (EVD) occurred simultaneously in South Sudan (then The Sudan) and the Democratic Republic of Congo (DRC, then Zaire) nearly forty years ago. Although the causative agent and how it was transmitted were initially unknown to those with the task of containing these outbreaks, only 602 people became infected, 431 died and  the outbreaks were controlled within three months. The latest WHO data record that the current epidemic, the first case of which occurred in Guinea in December 2013, has so far infected over four thousand people in five West African countries and caused over two thousand deaths, more than in all  previous outbreaks put together. In a separate outbreak in the DRC there is a total of 24 suspected cases and 13 people have died so far. The obvious question is why, in spite of the initial ignorance about the disease in 1976, was the epidemic contained so promptly when all our current knowledge is now failing to stop the disease spreading exponentially?
Having established that the Ebola virus was spread via contact with patients’ body fluids, the teams investigating the 1976 outbreaks closed  hospitals that were reusing their few needles, quarantined infected people and their contacts, and effectively disseminated  information on simple barrier  nursing of patients and safe burial practices. The 2014 outbreak, though, started in a much less isolated region where people regularly travel between countries using motor vehicles. Healthcare workers from outside were often distrusted, and local hospital staff had no previous experience of EVD. In addition Muslim law requires family members to wash their dead before burial. As the initial outbreak became an epidemic, control efforts were hampered by lack of suitable hospital facilities, basic equipment and staff, forcing patients to return home and infect others. And sadly some protective responses from the international community, such as closing airline routes, have greatly exacerbated the problem in West Africa by preventing medical experts and supplies, even food, from reaching affected areas.
The WHO now predicts that it will take 600 million dollars and at least six to nine months to control the epidemic. The European Union, United States and the Gates Foundation have all committed funds for vaccine development and treatment, but according to Medecins Sans Frontieres the international response is still “lethally inadequate”. The imperative need now is for experts in  biohazard containment.

Tumour markers have been widely used in clinical settings for early cancer detection, diagnosis, prognosis and recurrence surveillance. Due to the growing usage, it is of vital importance to assess the performance of common tumour markers on in-vitro diagnosis instruments. In this study, the most commonly used tumour markers have been selected to evaluate the performance of the SNIBE Maglumi 2000  chemiluminescence immunoassay system by comparing with our reference methods.

by Dr Xiao Hu, Dr Sheng Kang, Zhiyun Duan and Professor Guichen Zhang

Background
Tumour markers are substances that rise abnormally in the body when cancer is present. They are useful indicators for cancer risk determination, screening, diagnosis, prognosis, post-treatment surveillance and recurrence monitoring [1]. Alpha-fetoprotein (AFP) is a well established marker in liver cancer diagnosis and post-treatment monitoring [2]. Another well studied tumour marker, prostate-specific antigen (PSA), is recommended for the screening of prostate cancer with men over 50 years old [3]. Carcinoembryonic antigen (CEA) is particularly used as a tumour marker for bowel cancer. It measures the response to treatment and monitors whether the disease has revisited [4]. Elevated serum ferritin has been found in patients with pancreatic cancer, breast cancer, colon cancer, non small cell lung cancer, hepatocellular carcinoma and Hodgkin’s lymphoma [5]. Cancer antigen 125 (CA 125) is a marker commonly used for following up patients with ovarian cancer after treatment [6], while cancer antigen 15-3 (CA 15-3) is widely used for breast cancer management [7]. Cancer antigen 19-9 (CA 19-9) is the best validated marker for pancreatic cancer post-treatment evaluation [8]. Cytokeratin 19 fragment (CYFRA 21-1) and squamous cell carcinoma antigen (SCCA) are useful markers for lung cancer diagnosis in combination with other markers [9] [10]. This study has evaluated the performance of ten tumour markers on the SNIBE Maglumi 2000 chemiluminescence immunoassay system.

Precision
According to the principle and method of the CLSI EP5-A2 guideline [11], we made some adjustments to evaluate the precision of ten tumour markers. Intra-assay precision was evaluated on three different levels of serum samples. Each sample was repeatedly measured for 20 times in the same run to calculate the coefficient of variation (CV%). Inter-assay precision was assessed by repeatedly measuring three different levels of samples for 10 days with the same batch of kit. Samples were run in duplicates, two runs per day with at least 3 hours time interval to calculate the coefficient of variation. The results are displayed as mean value and CV%. Table 1 lists the precision results of ten tumour markers. The CVs of the intra- and inter-assays were less than 4.12% and 6.67% respectively (Table 1).

Method comparison
Serum samples from patients with benign diseases to various cancers were offered by the clinical laboratory of our hospital. The patient names were coded with confidentiality. The samples were measured by our reference system and the SNIBE Maglumi 2000 system to form correlation dot plots. Concordance between SNIBE Maglumi 2000 and reference systems for each tumour marker was analysed. For each tested marker, the number of serum samples is ranged from 166 to 460.

By comparing with our reference methods, good correlations were shown between the SNIBE Maglumi 2000 and the ROCHE Cobas e601 or the ABBOTT Architect i2000. The slopes for all markers were between 0.853 and 1.361 while the intercepts ranges from -2.515 to +5.138 (Figure 1A-J). Total PSA has the highest correlation between the SNIBE Maglumi 2000 and the ROCHE Cobas e601 while the lowest relevance (R2=0.981) was seen in CA 19-9 between the SNIBE Maglumi 2000 and the ROCHE Cobas e601 (Figure 1). The total coincidence rate is between 93.7% (Figure 1I) and 99.6% (Figure 1B).

Conclusion
In this study, we have evaluated the performance of the SNIBE Maglumi 2000 chemiluminescence immunoassay system via ten tumour markers. The intra-assay precision and inter-assay precision for all markers examined here are highly acceptable. By comparing with our reference methods, a high correlation has been shown for all markers tested with the SNIBE Maglumi 2000 system. The total coincidence rate is within the acceptable range for all markers examined. To conclude, the SNIBE Maglumi 2000 system is reliable for the measurement of tumour markers in clinical use.

References
1. Sturgeon CM, Hoffman BR, Chan DW, et al. National academy of clinical biochemistry laboratory medicine practice guidelines for use of tumour markers in clinical practice: Quality requirements. Clin Chem. 2008; 54 (8): e1-e10.
2. Manini MA, Sangiovanni A, Fornari F, et al. Clinical and economical impact of 2010 AASLD guidelines for the diagnosis of hepatocellular carcinoma. J Hepatol. 2014; 60 (5): p995-1001.
3. Qaseem A, Barry MJ, Denberg TD, et al. Screening for prostate cancer: a guidance statement from the clinical guidelines committee of the American College of Physicians. Ann Intern Med. 2013; 158 (10): 761-769.
4. Labianca R, Nordlinger B, Beretta GD, et al. Primary colon cancer: ESMO Clinical Practice Guidelines for diagnosis, adjuvant treatment and follow-up. Ann Oncol. 2010; 21 (S5): v70-v77.
5. Alkhateeb AA, Connor JR. The significance of ferritin in cancer: Anti-oxidation, inflammation and tumorigenesis. BBA. 2013; 1836 (2): 245-254.
6. Forstner R, Sala E, Kinkel K, et al. ESUR guidelines: ovarian cancer staging and follow-up. Eur Radiol. 2010; 20: 2773-2780.
7. Sandri MT, Salvatici M, Botteri E, et al. Prognostic role of CA15.3 in 7942 patients with operable breast cancer. Breast Cancer Res Treat. 2012; 132:317–326.
8. Duffy MJ, Sturgeon C, Lamerz R, et al. Tumor markers in pancreatic cancer: a European Group on Tumor Markers (EGTM) status report. Ann Oncol. 2010; 21: 441-447.
9. Molina R, Auge JM, Escudero JM, et al. Mucins CA 125, CA 19.9, CA 15.3 and TAG-72.3 as tumour markers in patients with lung cancer: comparison with CYFRA 21-1, CEA, SCC and NSE. Tumour Biol. 2008; 29:371–380.
10. Chu XY, Hou XB, Song WA, et al. Diagnostic values of SCC, CEA, Cyfra21-1 and NSE for lung cancer in patients  with suspicious pulmonary masses: A single center analysis. Cancer Biol Ther. 2011; 11(12): 995-1000.
11. Tholen DW, Kallner A, Kennedy JW, et al. Evaluation of precision performance of quantitative measurement methods; approved guidelines-second edition, EP5 A2. 2004; 24(25).

The authors
Xiao Hu* MD, Sheng Kang PhD, Zhiyun Duan MSc, Dept of Clinical Laboratory, Shenzhen Sixth People’s Hospital, Shenzhen, Guangdong 518052, China
Guichen Zhang Professor, PhD, MD, Medical College, Shenzhen University, Shenzhen, Guangdong 518052 China
(*Corresponding author:  xiao121386@163.com)

The human gene for anti-Müllerian hormone (AMH) was isolated and sequenced 20 years ago [1], with the first immunoassays developed in 1990 [2,3].  Since then, our understanding of this hormone has significantly increased, with most clinical use today focusing on women’s reproductive health. AMH’s ability to reflect the number of small antral and pre-antral follicles present in the ovaries, and therefore the ovarian reserve, has led to AMH measurement being used in a wide array of clinical applications.

One of the first was as a tumour marker in the diagnosis and follow up of women with ovarian granulosa cell tumours (GCT) [4, 5]. More recently, with the dramatic improvements in the treatment of childhood cancers, attention is focused on AMH to assess the likelihood of gonadal damage and infertility after treatment.  It is also being used to investigate the toxicity of different therapeutic regimens, in the choice of those treatments, and the prediction (and potential preservation) of fertility in young women and children following cancer therapy.

Sensitive diagnostic marker for GCT
GCT accounts for 2-3% of all ovarian tumours, with two distinct types: the juvenile and the adult form. The more common adult form generally presents in women at around 50 years. A majority have endocrine manifestations as a direct consequence of hormone secretion by the tumour [6].

GCTs have the potential to secrete estradiol, Inhibin (A and B) and AMH. Inhibin and AMH are the more useful biomarkers since estradiol is only produced in 50-60% of GCT patients and is dependent on stimulation by testosterone from adjacent theca cells. While serum total Inhibin is secreted in almost all GCT and has been shown to successfully detect recurrence following surgery, it is also increased in some epithelial ovarian tumours and fluctuates significantly within the menstrual cycle. AMH is more specific to GCT as expression is limited to ovarian granulosa cells and it does not change substantially over the menstrual cycle.

Although GCT is extremely rare, it is noted for its late recurrence, usually within four-six years, but can be up to 10-20 years after removal of the primary tumour. AMH disappears within days of removal of the ovaries [7] and, following tumour resection, a rise in AMH precedes clinical detection, making it an extremely sensitive marker for the early detection of tumour recurrence.

Lane’s 1999 study followed 56 patients post operatively and showed that AMH was useful in evaluating the completeness of tumour removal [4]. In addition, serial AMH measurements were able to detect recurrence on average three months prior to clinical detection. A second study, which followed 31 patients for up to seven years, confirmed these observations [5]. This group used an AMH assay 20 times more sensitive than previously used and, when comparing both assays found discrepant values in six out of 31 patients. The more sensitive assay accurately reflected the clinical situation and was elevated up to 16 months earlier in patients with tumour recurrence.

However, there is still insufficient published information on which to assess the sensitivity and specificity of AMH for the diagnosis of GCT.  This is due to small patient numbers, the insensitivity of older assays and the lack of solid reference values in pre-menopausal women and children. The advent of more sensitive, fully automated assays will facilitate more robust studies.

Assessment of ovarian damage
The relationship between AMH and the number of small growing follicles (and therefore the number of primordial follicles or ovarian reserve) makes it useful for assessing the gonadal toxicity of cancer therapy and loss of ovarian reserve.  Levels fall rapidly with the onset of cancer treatment, with subsequent recovery dependant on degree of ovarian damage. AMH appears to identify which treatments may spare the ovaries, or are most toxic to them, and may give clinicians additional information to direct therapeutic choices in children and women of childbearing age with cancer.

Radiotherapy is a well-known cause of ovarian damage, even at low radiation levels. Women who have undergone pelvic or total body irradiation are likely to have low or undetectable AMH levels [9, 10]. The gonadal toxicity of alkylating agents is also well established. In a study involving young women with lymphoma, those receiving alkylating agents showed little or no recovery in AMH levels following treatment whereas those receiving alternative chemotherapy showed good recovery. 

Childhood cancer and fertility
Childhood cancer treatment has improved dramatically with survival rates of more than 90%.  However, the consequences of treatment may be permanent damage to the ovaries, affecting fertility. AMH is detectable in females of all ages rising steadily throughout childhood. Several studies have confirmed its role as a clinically useful marker to assess impairment of ovarian reserve in those receiving treatment for cancer  [11, 12, 13].

Brougham showed that AMH decreased during chemotherapy in both prepubertal and pubertal girls, becoming undetectable in 50% of patients; recovery occurred in the low to medium risk groups after completion of treatment, yet remained undetectable in the high risk group.  Inhibin B was undetectable in most patients before treatment and FSH showed no relationship with treatment. Thus AMH indicates a more useful  assessment of residual ovarian reserve, revealing partial loss or ovarian failure.

It is clear that a woman can suffer a significant loss of ovarian reserve without any lasting effects on her fertility, for example following removal of an ovary.  For survivors of childhood cancer this may mean that only a substantial loss of ovarian reserve would have a clinical impact. Indeed, recent work has shown that there is a high number of successful pregnancies in lymphoma survivors, despite low AMH levels [14].  In a study of 84 childhood cancer survivors they achieved pregnancy rates similar to controls despite impaired ovarian reserve [15]. However, a 10-year follow up study of childhood cancer survivors, now in their 30s, showed that the percentage of childless women in this group was greater than in the normal Danish population, particularly in the group of women who received the most gonadotoxic treatment burden. Their pregnancy rate and outcome was especially poor [16].  The truth is difficult to discern on current evidence and more work is required on long term follow up, with fertility and age at menopause as end points.

The real value of measuring AMH in young women surviving cancer would be to forecast long-term reproductive outcome and take steps to preserve their fertility.

Reproductive outcomes in adult women
The same fertility concerns exist for women of childbearing age.  Using AMH values to assess ovarian reserve and individualize risk, more invasive methods of fertility preservation may be appropriate for women with a low AMH, while those with high values for their age may decide  to start cancer treatment without delay.

Most evidence comes from breast cancer studies and is based on the assumption that a woman with a higher pre-treatment AMH before chemotherapy will be more likely to retain ovarian function. A prospective study in women with newly diagnosed breast cancer linked high levels of AMH detected before treatment with retaining long-term ovarian function five years after surgery [17]. Pretreatment serum AMH was seen to be markedly higher in women who continued to have menses. The predictive value of AMH for post-chemotherapy ovarian function has subsequently been confirmed [18] allowing the development of prediction tools combining age and AMH [18].

Individualizing breast cancer adjuvant chemotherapy
Adjuvant endocrine therapy has been shown to reduce the likelihood of reocurrence and improve overall survival rates in hormone receptor-positive (HR-positive) breast cancer. However, it appears that ovarian function after chemotherapy has direct implications on the choice of therapy.  Aromatase inhibitors (AIs) are more effective in postmenopausal women than tamoxifen [19]. However, in premenopausal women, AIs may cause a rise in estrogen levels due to reactivation of ovarian function. Consequently, even in women who have developed chemotherapy-induced ovarian failure, tamoxifen is the standard of care [20, 21].

It has been suggested that all women who are premenopausal prior to chemotherapy, even those in their late 40s and early 50s, should be treated with adjuvant tamoxifen therapy or, if they are going to receive an aromatase inhibitor, should have their ovaries removed or chemically suppressed [22]. For the latter group, these strategies are invasive and are associated with increased side effects. Consequently, being able to predict permanent ovarian failure using information other than the patient’s age is relevant.

Data from recent studies [8, 17] suggest that  pre-chemotherapy assessment of serum AMH concentrations, possibly in combination with inhibin B, may provide important information about the likelihood of developing permanent ovarian failure with chemotherapy. In addition, this could help identify a patient population in which it would be safe to treat with upfront AI monotherapy. The expanding number of studies available all add to our understanding of the role of AMH in ovarian function, its ability to predict a woman’s ovarian reserve for her fertility and the impact of cancer treatment on reproductive health.

References
1. Cate RL, Mattaliano RJ, Hession C, et al.  Isolation of the bovine and human genes for Müllerian inhibiting substance and expression of the human gene in animal cells. Cell 1986; 45, 685-698.
2. Hudson PL, Dougas I, Donahoe PK, et al. An immunoassay to detect human Müllerian inhibiting substance in males and females during normal development. J Clin Endocrinol Metab. 1990; 70, 16-22.
3. Josso et al. An enzyme linked immunoassay for anti-müllerian hormone: a new tool for the evaluation of testicular function in infants and children. JCEM 1990; 70, 23-27.
4. Lane AH, Lee MM, Fuller AF Jr, et al. Diagnostic utility of Müllerian inhibiting substance determination in patients with primary and recurrent granulosa cell tumors. Gynecol Oncol 1999; 73, :51–55.
5. Long WQ, Ranchin V, Pautier P, et al. Detection of minimal levels of serum anti-Müllerian hormone during follow-up of patients with ovarian granulosa cell tumor by means of a highly sensitive enzyme-linked immunosorbent assay. J Clin Endocrinol Metab 2000; 85, 540–544.
6. Bjorkholm E, Silfversward C. Prognostic factors in granulosa-cell tumors. Gynecol Oncol. 1981;11, 261–274.
7. LaMarca A, De Leo V, Giulini S, et al. Anti-Müllerian hormone in premenopausal women and after spontaneous or surgically induced menopause. J Soc Gynecol Invest 2005; 12, 545–548.
8. Henry, NL, Xia R, Schott AF, McConnell D, et al. Prediction of Postchemotherapy Ovarian Function Using Markers of Ovarian Reserve. The Oncologist 2014; 19, 68–74.
9. Lie Fong S, Laven JS, Hakvoort-Cammel FG, et al. Assessment of ovarian reserve in adult childhood cancer survivors using anti-Mullerian hormone. Hum Reprod 2009;24, 982–990
10. Gracia CR, Sammel MD, Freeman E, et al. Impact of cancer therapies on ovarian reserve. Fertil Steril 2012; 97, 134–140 e131.
11. Bath LE, Wallace WH, Shaw MP, et al. Depletion of ovarian reserve in young women after treatment for cancer in childhood: detection by anti-Müllerian hormone, inhibin B and ovarian ultrasound. Hum Reprod 2003; 18, 2368–2374.
12. van Beek RD, van den Heuvel-Eibrink MM, Laven JS, et al. Anti-Müllerian hormone is a sensitive serum marker for gonadal function in women treated for Hodgkin’s lymphoma during childhood. J Clin Endocrinol Metab 2007; 92, 3869–3874.
13. Brougham MF, Crofton PM, Johnson EJ, et al. Anti-Müllerian hormone is a marker of gonadotoxicity in pre- and postpubertal girls treated for cancer: a prospective study. J Clin Endocrinol Metab 2012; 97, 2059–2067.
14. Janse F, Donnez J, Anckaert E, et al. Limited value of ovarian function markers following orthotopic transplantation of ovarian tissue after gonadotoxic treatment. J Clin Endocrinol Metab 2011; 96, 1136–1144.
15. Dillon KE, Sammel MD, Ginsberg JP, et al. Pregnancy After Cancer: Results From a Prospective Cohort Study of Cancer Survivors. Pediatr Blood Cancer. 2013 Dec; 60(12), 2001-6.
16. Nielsen SN, Andersen AN, Schmidt KT, et al. A 10-year follow up of reproductive function in women treated for childhood cancer. Reprod Biomed Online 2013; 27, 192–200.
17. Anderson RA, Cameron DA. Pretreatment serum anti-müllerian hormone predicts long-term ovarian function and bone mass after chemotherapy for early breast cancer. J Clin Endocrinol Metab 2011; 96, 1336–1343.
18. Anderson RA, Rosendahl M, Kelsey TW, et al. Pretreatment anti-Müllerian hormone predicts for loss of ovarian function after chemotherapy for early breast cancer. Eur J Cancer 2013;49, 3404–3411.
19. Burstein HJ, Prestrud AA, Seidenfeld J et al. American Society of Clinical Oncology clinical practice guideline: Update on adjuvant endocrine therapy for women with hormone receptor-positive breast cancer. J Clin Oncol 2010; 28, 3784–3796.
20. Smith IE, Dowsett M, Yap Y-S et al. Adjuvant aromatase inhibitors for early breast cancer after chemotherapy-induced amenorrhoea: Caution and suggested guidelines. J Clin Oncol 2006; 24, 2444–2447.
21. Burstein HJ, Mayer E, Patridge AH et al. Inadvertent use of aromatase inhibitors in patients with breast cancer with residual ovarian function: Cases and lessons. Clin Breast Cancer 2006;7, 158–161.
22. Henry NL, Xia R, Banerjee M et al. Predictors of recovery of ovarian function during aromatase inhibitor therapy. Ann Oncol 2013; 24, 2011–2016.

The author
Sherry Faye, PhD
Director, Global Scientific Affairs,
Beckman Coulter Diagnostics
Brea, CA, USA