A 5-day-old girl died suddenly and unexpectedly just after waking up from a brief nap. The pregnancy was normal, the baby was born at term, and the delivery was uneventful. There had been no signs of infection, and the baby had appeared healthy. The autopsy revealed diffuse liver steatosis. Postmortem measurements of acylcarnitines in spotted blood and bile samples were performed using tandem mass spectrometry. The postmortem metabolic analysis revealed abnormally high levels of several acylcarnitine species in both the blood (Figure 1) and bile (Figure 2). In particular, the concentration of octanoylcarnitine (C8) was markedly elevated in both sample types.

Figure 1: Acylcarnitine profile in dried blood spot measured with tandem mass spectrometry
Figure 2: Acylcarnitine profile in bile measured with tandem mass spectrometry 

Considering the acylcarnitine results, what is the most likely cause of death?

  • Glutaric aciduria type II
  • Medium-chain acyl-CoA dehydrogenase deficiency
  • Very-long-chain acyl-CoA dehydrogenase deficiency
  • Carnitine uptake defect

The correct answer is ...

Carnitine uptake defect

Several metabolic disorders can present with sudden unexpected childhood death. These include mitochondrial β-oxidation disorders, mitochondrial respiratory-chain disorders, and several organic acidemias. Many of these metabolic disorders can be detected postmortem by examining acylcarnitines in spotted blood and bile samples.

Carnitine uptake defect (CUD)

Carnitine is essential for the transport of long-chain fatty acids over the mitochondrial inner membrane. CUD is an autosomal recessive disease caused by defects in the organic cation/carnitine transporter (OCNT2) — a high-affinity carnitine transporter encoded by the gene SLC22A5. OCNT2 is responsible for carnitine uptake in the kidney, heart, muscle, and fibroblast but not in the liver. Defects of the carnitine transporter impair renal reabsorption and muscle uptake of carnitine and, consequently, cause primary carnitine deficiency, characterized by extremely low plasma carnitine concentration and impairment of mitochondrial fatty acid oxidation. 

Symptoms may present in infancy (usually before 2 years of age) with hypoglycemia, liver dysfunction, and hyperammonemia. Sudden and unexpected deaths have been reported. Alternatively, symptoms may present later in childhood with progressive cardiomyopathy, often accompanied by skeletal muscle weakness, or in adulthood with fatigue and arrhythmias. 

Acylcarnitine analysis of dried blood spots from untreated individuals with CUD shows abnormally low levels of free carnitine and acylcarnitines (low total carnitine). 

Glutaric aciduria type II (GA-II)

GA-II is caused by the inability to transfer electrons from multiple flavoprotein acyl-CoA dehydrogenases to the respiratory chain. These enzymes are involved in various metabolic pathways, including fatty acid β-oxidation and amino acid catabolism. The condition is consequently also known as multiple acyl-CoA dehydrogenase deficiency. 

The metabolic defects are, most commonly, in the electron transport flavoprotein (ETF) and the ETF-ubiquinone oxidoreductase, which together transfer electrons from mitochondrial FAD-linked dehydrogenases to the ubiquinone pool of the respiratory chain. GA-II also can result from defects of riboflavin transport and processing. 

Symptoms may present at any age and vary widely in severity. The severe, neonatal-onset form presents with hypoketotic hypoglycemia, metabolic acidosis, and hyperammonemia in the first days of life. Congenital anomalies may be present, including cystic kidneys, facial dysmorphism, and neuronal migration defects. Many affected individuals die within a week of birth. Those that survive the neonatal period may develop cardiomyopathy and die within months. The milder form can present at any age and is most often associated with muscle weakness, exercise intolerance, and muscle pain. 

Acylcarnitine analysis of blood spots from individuals with GA-II typically shows increased levels of multiple acylcarnitines, from C4 to C16, reflecting the various metabolic pathways affected. 

Medium-chain acyl-CoA dehydrogenase (MCAD) deficiency

MCAD is a FAD-linked dehydrogenase that catalyzes the first step of the mitochondrial β-oxidation spiral, i.e., the oxidation of the α-β carbon bond of acyl-CoAs, with a high affinity towards medium-chain length fatty acids (C6-C10).

MCAD deficiency is an autosomal recessive disorder caused by defects of the gene ACADM, which leads to impaired mitochondrial oxidation of medium chain-length fatty acids. Affected individuals appear normal until an episode of acute metabolic decompensation is provoked by a period of fasting, often concurrent with an infection. The first episode of acute metabolic decompensation may occur at any age — from the neonate period to adulthood. A small percentage develop severe, life-threatening symptoms during the first week of life, at times, before the result from newborn screening are available. Before widespread newborn screening for MCAD deficiency, affected individuals typically presented between 3 months and 24 months with hypoketotic hypoglycemia, encephalopathy, and liver dysfunction. Unexpected death during the first metabolic decompensation was common. 

Acylcarnitine analysis of dried blood spots and bile samples from MCAD-deficient individuals show elevated levels of medium chain-length acylcarnitines (C6-C10), most often with the characteristic pattern of C6<C8>C10.

Very-long-chain Acyl-CoA dehydrogenase (VLCAD) deficiency 

VLCAD is a FAD-linked dehydrogenase that catalyzes the first step of mitochondrial β-oxidation with an affinity toward long-chain fatty acids (C14-C20).

VLCAD deficiency is an autosomal recessive disorder caused by a defect in the gene ACADVL, which leads impaired mitochondrial oxidation of long chain-length fatty acids. Three distinct phenotypes have been observed: a severe neonatal form that presents with cardiomyopathy, hypotonia, and hepatomegaly; a childhood-onset form that presents with hypoketotic hypoglycemia and hepatomegaly, but without cardiomyopathy; and a late-onset form that presents with exercise- or illness-induced rhabdomyolysis, muscle weakness, and exercise intolerance. VLCAD deficiency may result in unexpected and sudden death if not detected and treated. 

Acylcarnitine analysis of dried blood spots and bile samples from VLCAD deficient individuals show elevated levels of long chain-length acylcarnitines (C14-C18), particularly C14:1.


  1. Bennett MJ, Rinaldo P. The metabolic autopsy comes of age. Clin Chem. 2001;47(7):1145-6. Epub 2001/06/28.
  2. Morris AAM, Spiekerkoetter U. Disorders of Mitochondrial Fatty Acid Oxidation & Riboflavin Metabolism. In: Saudubray J-M, Baumgartner MR, Walter J, editorsInborn Metabolic Diseases: Diagnosis and Treatment. Berlin, Heidelberg: Springer Berlin Heidelberg; 2016. p. 201-13.
  3. Rinaldo P, Matern D, Bennett MJ. Fatty acid oxidation disorders. Annu Rev Physiol. 2002;64:477-502. Epub 2002/02/05.
  4. Rinaldo P, Stanley CA, Hsu BYL, Sanchez LA, Stern HJ. Sudden neonatal death in carnitine transporter deficiency. The Journal of Pediatrics. 1997;131(2):304-5.
  5. Angle B, Burton BK. Risk of sudden death and acute life-threatening events in patients with glutaric acidemia type II. Mol Genet Metab. 2008;93(1):36-9. Epub 2007/11/03.
  6. Coughlin CR, 2nd, Ficicioglu C. Genotype-phenotype correlations: sudden death in an infant with very-long-chain acyl-CoA dehydrogenase deficiency. J Inherit Metab Dis. 2010;33 Suppl 3:S129-31. Epub 2010/01/29.

Freyr Johannsson, Ph.D.

Resident, Clinical Biochemical Genetics
Mayo Clinic

Silvia Tortorelli, M.D., Ph.D.

Consultant, Biochemical Genetics
Mayo Clinic
Associate Professor of Laboratory Medicine and Pathology
Mayo Clinic College of Medicine and Science

You are asked to test a computer vision application created to detect Anomaly X on digital pathology slides. Anomaly X is considered to be a rare event. You are provided the following:

Model Type: Logistic Regression Image Classifier

Images in training set: 56 positive for Anomaly X, 200 negative for Anomaly X

Accuracy of detecting Anomaly X on training set: 0.95

When you attempt to validate the application on 50 of your own digital pathology slides that are known to have anomaly X and 50 slides without, the application is only able to correctly identify 20% of the slides as positive.

Which of the following could explain why the application failed to perform with high accuracy on your test set of slides?

A. The machine learning model has been overfitted to the training images.

B. Cross-validation and subsequent hyperparameter tuning was not performed on the training data.

C. Logistic regression is not a valid approach for image classification.

D. Choices A and B.

The correct answer is ...

Choices A and B.

Ensuring that a machine learning model trained on a specific set of data is adaptable to real world data, also known as generalization, is important. Data overfitting is a common issue that is seen with models that have excellent performance with training data but performs poorly on test data. Overfitting is when the model “learns” the unique characteristics of the data it was trained on, rather than on characteristics that would be more broadly applicable to its intended task.

The green line represents a model that is overfitted on the training data to classify between red and blue dots. The black line represents a model that has optimal fit on classifying the two. (Image adapted from https://en.wikipedia.org/wiki/Overfitting#/media/File:Overfitting.svg, licensed under BY-SA 4.0).

Common approaches to prevent overfitting include repeatedly training the model on different slices of the training set, while also leaving out a subset to test the accuracy of the model. This is known as cross-validation. Subsequent adjustment of the model after cross-validation can further help the model’s performance. This is also known as hyperparameter tuning. 

Other ways to prevent overfitting include simplifying or decreasing the number of features being used for the model, also known as regularization, and increasing the amount of data available for the model to be trained on.

Choice C is incorrect; logistic regression is a very popular machine learning method for classification, including image classification. 


  1. Glassner, Andrew S. “Chapter 9: Overfitting and Underfitting.” Deep Learning: A Visual Approach, No Starch Press, Inc, San Francisco, CA, 2021.
  2. Zheng, Alice. Evaluating Machine Learning Models, O'Reilly Media, Inc., Sebastopol, CA, 2015.

Ray Qian, M.D.

Fellow, Clinical Informatics
Mayo Clinic

Photo of Chady Meroueh, M.D.

Chady Meroueh, M.D.

Senior Associate Consultant, Anatomic Pathology
Mayo Clinic

A 32-year-old man presented with a 2.8 cm testicular tumor. On gross examination, a hemorrhagic and ill-defined nodular area with a friable cut surface within the testicular tissue was noted. The surface of the tunica vaginalis was smooth and unremarkable except for the area where the hemorrhagic nodule was present.

Figure 1: ER
Figure 2: PAX8
Figure 3: Serous 1
Figure 4: Serous 2
Figure 5: Serous 3

Based on the history and photomicrographs, which one of the following diagnoses is correct?

  • Postpubertal-type teratoma
  • Adenocarcinoma of the rete testis
  • Mesothelioma
  • Serous borderline tumor

The correct answer is ...

Serous borderline tumor

Our case demonstrates a neoplasm with branching papillary architecture associated with cystic spaces. The tumor is lined by stratified serous-type epithelium with mild to moderate atypia and psammomatous calcifications. There is no stromal invasion. The neoplastic cells are positive for PAX8, WT1, and ER, which support the diagnosis of a serous borderline tumor.

The differential diagnosis includes mesothelioma, which can show microcystic and micropapillary patterns as seen in a serous borderline tumor. However, the overall immunophenotype rules out this possibility.

Another important differential diagnosis is teratoma, but other teratomatous components are not present. Also, no morphological evidence of germ cell neoplasia in-situ (GCNIS) was identified in the background seminiferous tubules, and this was supported by the absence of staining for OCT3/4.

Adenocarcinoma of the rete testes is a high-grade malignancy arising from the rete testis in the hilar region. These tumors show solid, tubular, or papillary patterns with clear stromal invasion, which is not evident in our case. Importantly, these tumors can be positive for PAX8; however, unlike serous borderline tumors, these tumors lack ER positivity.

Serous borderline tumors of testes are rare, and in one study, only 5 such cases were identified among 146 paratesticular and rete testis tumors.(1) Serous tumors of the testes, although rare, are the most common subtype of mullerian epithelial tumors and can show the same spectrum of benign, borderline, and invasive carcinomas as seen in the ovary.(2)

Serous borderline tumor of the testes are frequently located at paratesticular sites. However, intratesticular tumors can also be seen as in the case highlighted herein. These tumors are rare and have an indolent clinical course. Stromal invasion should be ruled out, as it carries adverse prognostic significance.(3)


  1. Appendageal tumors and tumor-like lesions of the testis and paratestis: a 32-year experience at a single institution. https://doi.org/10.1016/j.humpath.2020.06.006
  2. Geramizadeh B, Farzaneh MR, Pakbaz S, Zeighami S. Testicular papillary serous cystadenocarcinoma: a rare case report and review of the literature. Rare Tumors. 2011;3(4):e44. doi:10.4081/rt.2011.e44
  3. AMAR S. IBRAHIM, CHENG LI and MOHAMMAD S. AL-JAFARI Borderline Serous Papillary Tumour of the Testis: A Case Report and Review of the Literature Anticancer Research. November 2012, 32 (11) 5011-5013.

Rabia Zafar, M.D.

Fellow, Surgical Pathology
Mayo Clinic

Sounak Gupta, M.B.B.S., Ph.D.

Senior Associate Consultant, Anatomic Pathology
Mayo Clinic
Assistant Professor of Laboratory Medicine and Pathology
Mayo Clinic College of Medicine and Science

A 27-year-old presented with papular skin rash in the inguinal region and palpable right inguinal lymphadenopathy. Excisional biopsy of right inguinal lymph node was performed and the histopathology is shown in Figure 1.

Figure 1

What is your diagnosis?

  • Langerhans cell histiocytosis
  • Dermatopathic lymphadenopathy
  • Rosai-Dorfman disease
  • Erdheim-Chester disease

The correct answer is ...

Dermatopathic lymphadenopathy

Dermatopathic lymphadenopathy is characterized by paracortical expansion by small lymphocytes, histiocytes, and Langerhans cells.(1) The sinuses are patent. Pigment laden macrophages (arrow, Figure 1) can be seen. CD1a marks the paracortical expansion by the Langerhans cell in a similar distribution to CD3-positive T-lymphocytes in the paracortex.

Rosai-Dorfman disease (RDD) is a non-Langerhans cell histiocytosis also known as sinus histiocytosis with massive lymphadenopathy. Morphologically, RDD is characterized by emperipolesis (engulfment of inflammatory cells by large histiocytes) with expression of CD68, CD163, S100, and negative for CD1a and langerin.

Langerhans cell histiocytosis (LCH) is characterized by infiltration of Langerhans cells restricted to the lymph node sinuses.(1) The pattern of distribution of Langerhans cells is the key morphologic feature to distinguish LCH from dermatopathic lymphadenopathy.

Erdheim-Chester disease (ECD) is a non-Langerhans cell histiocytosis and a clonal systemic process with multiorgan involvement and variable clinical outcomes. Typical clinical presentation of ECD includes central diabetes insipidus, perinephric fibrosis, and sclerotic bone lesions. The lesional histiocytes strongly express CD68, CD163, Factor XIIIa, and are negative for CD1a and langerin. BRAF V600E is positive in a subset of cases.


  1. Ravindran A, Goyal G, Failing JJ, Go RS, Rech KL. Florid dermatopathic lymphadenopathy-A morphological mimic of Langerhans cell histiocytosis. Clin Case Rep. 2018 Jun 22;6(8):1637-1638. doi: 10.1002/ccr3.1663. PMID: 30147924; PMCID: PMC6099043.

Aishwarya Ravindran, M.B.B.S.

Resident, Hematopathology
Mayo Clinic

Photo of Karen Rech, M.D.

Karen Rech, M.D.

Consultant, Hematopathology
Mayo Clinic
Associate Professor of Laboratory Medicine and Pathology
Mayo Clinic College of Medicine and Science

A 73-year-old man with biventricular diastolic heart failure, spinal stenosis, chronic kidney disease, cirrhosis, hypertension, and bilateral carpal tunnel syndrome, was referred to the clinic for suspected cardiac amyloidosis. An echocardiogram revealed an abnormal global averaged left ventricular longitudinal peak systolic strain at −7% (normal = ≤−18%), apical sparing pattern. Green birefringence on Congo red-stained cardiac tissue was observed under cross-polarized light (A). Liquid chromatography tandem mass spectrometry on peptides extracted from Congo red-positive, microdissected areas of the specimen detected a peptide profile consistent with ATTR (transthyretin)-type amyloid deposition (B). However, Sanger sequencing of the TTR gene did not reveal a pathogenic missense alteration (C). 

Figure 1: A
Figure 1: B
Figure 1: C

What is the most likely diagnosis and/or appropriate follow-up for this patient?

  • Sanger sequencing data confirms this patient does not have an ATTR (transthyretin)-type amyloidosis.
  • Based on Congo red stain and mass spectrometry alone, we can conclude this patient has familial ATTR (transthyretin)-type amyloidosis TTR-associated.
  • Sanger sequencing is unable to identify the most common pathogenic TTR variants, and follow-up gene-targeted deletion/duplication analysis should be recommended.
  • Histology, mass spectrometry, and sequencing results are consistent with a diagnosis of wild-type ATTR (transthyretin)-type amyloidosis.

The correct answer is ...

Histology, mass spectrometry, and sequencing results are consistent with a diagnosis of wild-type ATTR (transthyretin)-type amyloidosis.

Correct. Congo red positivity on cardiac specimen indicates the patient does have amyloid fibrils in the heart. Mass spectrometry analysis confirms these are TTR-type amyloid fibrils. A negative Sanger sequencing result is most consistent with wild-type or age-related TTR-type amyloidosis. 

Sanger sequencing data confirms this patient does not have an ATTR (transthyretin)-type Amyloidosis. Incorrect — The negative Sanger sequencing results simply provides evidence that this patient is less likely to have an inherited or familial form of Transthyretin (TTR)-type Amyloidosis. Congo red positivity and mass spectrometry indicate that this patient does have TTR-type cardiac amyloidosis. 

Based on Congo red stain and mass spectrometry alone we can conclude this patient has Familial ATTR (transthyretin)-type Amyloidosis TTR-associated. Incorrect — Congo red stain positivity and the mass spectrometry results indicate that this patient does have TTR-type cardiac amyloidosis but does not specify familial or wild-type (age-related) forms.(1)

Sanger sequencing is unable to identify the most common pathogenic TTR variants and follow up gene-targeted deletion/duplication analysis should be recommended. Incorrect — More than 99% of pathogenic TTR variants are sequence variants, generally detectable by standard Sanger sequencing analysis. Gene-targeted deletion duplication analyses are extremely low yield, as these types of alterations (i.e., exon or whole-gene deletions/duplications) are not typically reported in familial TTR-type amyloidosis.(2)  


  1. Ashutosh D Wechalekar, A.D., et al. Review Lancet: Systemic amyloidosis. 2016 Jun 25;387(10038):2641-2654. doi: 10.1016/S0140-6736(15)01274-X. Epub 2015 Dec 21.
  2. Adam MP, Ardinger HH, Pagon RA, et al., editors. GeneReviews: Hereditary Transthyretin Amyloidosis. Seattle (WA): University of Washington, Seattle; 1993-2021.

Laura Thompson, Ph.D.

Fellow, Laboratory Genetics and Genomics
Mayo Clinic

Linda Hasadsri, M.D., Ph.D.        

Consultant, Laboratory Genetics and Genomics
Mayo Clinic
Assistant Professor of Laboratory Medicine and Pathology
Mayo Clinic College of Medicine and Science

MCL Education

This post was developed by our Education and Technical Publications Team.