March 2005
ANATOMIC
PATHOLOGY
Histopathology of
Liver Cancers
Kojiro M.
Recently, new pathomorphologic information about
early-stage small hepatocellular carcinoma (HCC) and
the multi-step process of human hepatocarcinogenesis
has been obtained, along with advances in the development of diagnostic
modalities. The most valuable information is that in the majority of cases HCC
arises as a very well differentiated cancer and proliferates with a stepwise
process of dedifferentiation. In addition, it has been suggested that many HCCs seem to arise from dysplastic
nodules (DNs) on the basis of the following evidence:
the presence of DNs containing HCC foci, frequent
association of DNs in the vicinity of HCC, and
clinical progression from DN to HCC. However, as many HCCs
are still detected at an advanced stage, it is also important to understand not
only the classical pathologic features of HCC but also unusual features such as
scirrhous change, sarcomatous
change, fibrolamellar variant, and intra-bile duct or intra-atrial tumor growth.
Recommendations
for the Reporting of Tissues Removed as Part of the Surgical Treatment of
Malignant Liver Tumors
The Association of Directors of Anatomic and Surgical Pathology (ADASP) has organized several committees to develop recommendations for the proper handling of surgical specimens and the preparation of the completed surgical pathology report. A committee of individuals with special interest and expertise in liver pathology has presented recommendations that were reviewed and approved by the ADASP council and subsequently by the membership.
The recommendations reported herein for primary and metastatic epithelial tumors in the liver are divided into 4 major areas: (1) informative gross description parameters, (2) diagnostic microscopic features recommended to be included in every surgical pathology report, (3) optional ancillary features that might be included in the final report, and (4) synoptic/checklist.
The checklist is generated from published studies on the diagnostic features and prognosis for primary and metastatic epithelial tumors in the liver. This checklist focuses on the most common primary epithelial tumors of the liver, hepatocellular carcinoma and cholangiocarcinoma. Also, this checklist emphasizes the handling of liver specimens from resections performed for metastatic colon adenocarcinoma, although the checklist could be used for other metastatic carcinomas as well.
The most widely adopted classification of liver-cell carcinoma is that proposed by the World Health Organization with added modifications. The grading of liver tumors described by Goodman and Ishak is a system in which hepatocellular carcinomas are separated into 4 grades, I to IV, on the basis of histologic differentiation. Grade I is the best differentiated, consisting of relatively small hepatocyte cells arranged in thin trabeculae. Grade II tumors consist of hepatocytes that are larger with abnormal nuclei and eosinophilic cytoplasm, and glandular structures might be seen. In grade III, giant tumor cells are more numerous. The cells of grade IV tumors are the most poorly differentiated, with hepatocytes possessing hyperchromatic, pleomorphic nuclei and little cytoplasm with loss of the trabecular pattern. Most hepatocellular carcinomas are grade II or III, with diagnostic difficulties most often occurring in the evaluation of well-differentiated (grade I) and poorly differentiated (grade IV) hepatocellular carcinoma.
The pathologic staging classification follows that of the pTNM from the International Union Against Cancer.
Features Recommended to Be Included in the Final Report
The surgical pathology report for partial or total liver resections should incorporate the following information.
A. General
1. How the specimen was identified: eg, labeled with name, medical record number, surgical pathology number
2. How the specimen was received: fresh or in fixative
3. The type of surgical procedure: segmentectomy, trisegmentectomy, partial lobectomy, complete resection
4. The anatomic site of the tumor in the liver
B. Gross Specimen
1. Weigh the specimen and give the dimensions in length width thickness.
2. Measure and describe the lesion(s).
a. Mark the resection margin with ink.
b. Description of the lesion(s).
Measure the lesion(s) in its greatest dimensions.
Is the lesion single or multiple, superficial or deep?
Does it involve the liver capsule, hepatic vein, portal vein, or inferior vena cava?
Are the extrahepatic biliary ducts and/or the hilum invaded?
(Can the involved liver segment be identified?)
c. Measure the distance between the inked resection margin and the nearest lesion.
d. Is there gross evidence of preexisting liver disease, eg, cirrhosis, hepatitis, hemochromatosis?
e. Is there evidence of prior chemoembolization, radiofrequency ablation?
f. Is there evidence of locoregional lymph node metastasis?
g. Is the gallbladder attached? Describe it.
C. Microscopic
Evaluation
1. State whether the tumor is a primary hepatocellular or cholangiocarcinoma or is metastatic. Describe microscopic peritumoral satellites, if present. A satellite is defined as a tumor nodule in the same segment or less than 2 cm from a lesion and less than 50% the diameter of the larger lesion and less than 4 cm in size, even if associated with a large mass.
2. State the grade of the tumor.
3. Document lymphovascular invasion, if present.
4. Document whether the resection margin is free of tumor and how closely the tumor is to the inked resection margin.
5. Describe lymph node involvement or lack thereof and the site of the lymph nodes: hilar nodes, celiac nodes, or juxtaregional (periaortic-pericaval/other intra-abdominal).
6. Describe foci of small cell dysplasia if possible.
7. Document underlying hepatic disease, if present, and state the type, eg, cirrhosis, with determination of etiology (eg, hepatitis B, hemosiderosis, a1-antitrypsin).
Does It Have a
Role in Diagnosis and Management?
Steatohepatitis, in particular nonalcoholic steatohepatitis (NASH), is one of the most common diagnoses made in routine liver biopsy practice, yet there remain gaps in our knowledge. Among these are basic questions, including the minimal histologic criteria for the distinction of steatosis from steatohepatitis, criteria for the prediction of outcome, and features allowing determination of cause or origin. The latter is an issue because steatohepatitis encompasses a wide spectrum of causes that fall into the general categories of toxic agents (alcohol and various medications) and metabolic disease, including most commonly the metabolic syndrome, although other metabolic diseases (eg, Wilson disease or tyrosinemia) technically might be included as well. These 3 aspects of steatohepatitisdiagnosis, prognosis, and cause or originare not independent because the criteria used for diagnosis may portend the prognosis and cause, and, conversely, different etiologic agents will determine the histologic features and prognosis, the latter perhaps independent of histologic features.
So what is the problem with diagnosis? NASH originally was defined by the clinical observation that liver disease with a histologic picture similar to alcoholic liver disease was occurring in patients who did not consume alcohol. Although somewhat controversial in its early days, epidemiologic considerations, including the association with obesity, diabetes, or both, soon persuaded most that NASH was a real entity. Given that realization, clarification of the histologic features of NASH vs alcoholic steatohepatitis (ASH) should have been possible because we could define the disease by epidemiologic rather than histologic features (because we were no longer constrained by the original definition of NASH having features similar to those of ASH). Defining a disease as "similar" to alcoholic liver disease always was limited by the presupposition that there was only 1 histologic appearance to alcoholic liver disease, which is far from true. Alcohol causes a wide range of histologic changes, ranging from simple macrovesicular steatosis to microvesicular steatosis in the form of acute foamy degeneration. So the histologic definition of "looking like alcoholic liver disease" has been problematic from the start.
The upshot of this is that the histologic description of NASH has included a wide range of features, and, in the defining article by Ludwig et al, simple fatty change without inflammation was included as a form of NASH. There is no question that the definition of steatohepatitis starts with fatty change as a necessary feature. Most current definitions of steatohepatitis require more than just fatty change, however, based on a study of a relatively small number of patients that often is quoted to demonstrate that the disease in patients with fatty change alone (or fatty change with inflammation) does not progress to cirrhosis, whereas in patients with ballooning degeneration, Mallory hyaline, and/or pericellular fibrosis, the disease progresses in a considerable number of cases.
Observational experience demonstrates that this rigid dichotomy is not absolutely true, but rather represents a relative truth and can be explained by viewing the results as indicating that we are looking at sequential steps of an ongoing process, which would fit the 2-hit hypothesis commonly invoked for steatohepatitis. This theory suggests that fatty change itself (steatosis) is relatively innocuous but primes the liver for a second hit, leading to more significant hepatocellular damage and, eventually, fibrosis (steatohepatitis). Livers with fatty change alone are livers waiting for the second hit, which may or may not come, whereas livers with more significant damage have already encountered their second hit and, therefore, are on a faster track to fibrosis.
Understanding this stepwise
progression from steatosis and steatohepatitis
is important in understanding this disease: the stepwise progression predicts
that intervention at the stage of steatosis alone
could prevent the second hit from occurring and that some livers demonstrating
only steatosis in a current biopsy specimen
eventually will have a second hit, and, therefore, disease will progress,
despite inferences that this does not happen. Nevertheless, as a practical
matter, livers with ballooning degeneration, Mallory hyaline, and/or fibrosis
have demonstrated that they have had the second hit and, therefore, are more
deserving of more intense follow-up. For this reason, the concept of diagnosing
steatohepatitis only after these changes are present
is a useful and practical one, as long as a diagnosis of steatosis
alone carries a disclaimer that the absence of features of steatohepatitis
does not guarantee that steatohepatitis will not
occur in the future in that patient.
The wide spectrum of histologic patterns seen with steatohepatitis
often is not appreciated when trying to distinguish different causes. For
example, amiodarone is a medication often reported as
causing "pseudoalcoholic liver disease."
However, my experience with the toxic effects of amiodarone
demonstrates that in many cases there is only mild fatty change, with extensive
and well-formed Mallory hyaline that seems out of proportion to the degree of steatosis. Marked cholestasis
also is common, again out of proportion to the other changes. These cases
usually look quite different from alcoholic liver disease or NASH, although
descriptively they might sound quite similar.
A similar situation exists in the
more common problem of distinguishing ASH from NASH related to the metabolic
syndrome. Several articles have highlighted histologic
difference, including differing types of inflammatory infiltrate, different
character to the Mallory hyaline, and other more unique changes like abundant glycogenated nuclei in NASH and more prominent central
venous changes with endophlebitis in severe cases of
ASH. Despite this, the differences sometimes are subtle or matters of degree,
and the article in this issue of the Journal by Sanderson and Smyrk applies a novel approach to distinguishing these
2 causes of steatohepatitis in a more direct manner.
The article by Sanderson and Smyrk attempts to distinguish NASH from ASH by using immunohistochemical analysis, making use of the known
relationship between insulin resistance and NASH. Liver biopsy specimens were
stained for insulin receptors (IRs) and for protein
tyrosine phosphatase 1B (PTP1B), a protein that acts
as a negative regulator of IR expression. By using the concept that cases of
obesity-related NASH might have decreased IR and increased PTP1B expression vs more normal IR expression and lower levels of PTP1B
expression in alcoholic liver disease, cases were categorized immunohistochemically into NASH and ASH and compared with
the clinical diagnosis made in each case.
In general, authors hypothesis was validated; there was reasonably good correlation between immunohistochemical results and the clinical diagnosis; however, the correlation was not perfect, with a number of clinical ASH cases staining as NASH (16/53 [30%]) and a number of clinical NASH cases staining as ASH (23/188 [12.2]%). As the authors discuss, there is nothing exclusive about the diagnoses of NASH and ASH. Having a NASH-like staining pattern in the ASH group might be explained by a combination of obesity-related insulin resistance in a group of patients with alcohol-related liver injury, the NASH-like staining pattern being the more specific of the 2 patterns (ie, the "ASH" pattern is, in reality, a normal staining pattern of liver tissue, not a specific pattern related to alcohol use). To support this contention, the authors point out that the body mass index of patients with NASH-staining clinically diagnosed ASH was slightly higher than in the clinically diagnosed ASH group as a whole (28.4 vs 27.2 kg/m2), although the difference is small. Explaining the lack of NASH-like staining in the clinical NASH group is a bit more difficult, unless the clinical diagnosis of NASH was simply incorrect or the underlying hypothesis about the relationship of IR expression to NASH is incomplete and patients with NASH might have different mechanisms of disease causation. These findings also could result from insensitivity of the method in patients with early disease, as suggested by the authors. The data presented do not allow us to make these distinctions.
Reference: Sanderson SO , Smyrk TC, Am J Clin Pathol 123:503-509, 2005
Editorial :
Am J Clin Pathol 123: April
2005
Expression of
cell adhesion molecules in oesophageal carcinoma and
its prognostic value
REVIEW
K S Nair ,
R Naidoo and R Chetty
Oesophageal carcinoma remains a disease of poor
prognosis. Surgical cure rates are compromised by the fact that most
patients are diagnosed at a late stage of disease because of the
delayed onset of symptoms, by which time metastases and organ
infiltration may have already occurred. Thus, invasion and
metastases play a key role in influencing patient survival, and the
search for novel treatments may therefore hinge on gaining insight
into the mechanisms controlling these processes. It has been
established that the initial step in the metastatic
cascade is the detachment of tumour cells
from the primary tumour via dysregulation
of normal cellcell and cellmatrix interactions. Distinct proteins
known as cell adhesion molecules (CAMs) mediate these
interactions. In recent years, a plethora of information has contributed
to the in depth understanding of these molecules. This review
provides a brief description of five families of CAMs
(cadherins, integrins,
CD44, immunoglobulin superfamily, and selectins) and highlights their altered expression in
relation both to prognosis and tumour behaviour in squamous cell
carcinoma and adenocarcinoma of the oesophagus.
Oesophageal
carcinoma is known for its extremely aggressive clinical behaviour and, despite improvement in surgical intervention
and preoperative management, the overall prognosis for patients is
poor. The assessment of prognosis through clinicopathological
characterisation remains inadequate using
standard grading and staging systems because of the considerable
variability and heterogeneity within different tumours
and stages.
"Elucidating the mechanisms controlling
invasion and metastasis in oesophageal
carcinoma may greatly assist in identifying those patients at higher
risk of metastases and mortality"
Currently, staging of the disease
is the most crucial parameter for predicting survival and
recurrence. Patients with minimal tumour
invasion of the oesophageal wall and without
metastases at the time of resection have a significantly better
chance of survival than those with lymph node metastases or organ
infiltration. This implies that the ability of tumour
cells to survive and grow at metastatic
sites considerably increases the level of morbidity. Therefore,
elucidating the mechanisms controlling invasion and metastasis in oesophageal carcinoma may greatly assist in
identifying those patients at higher risk of metastases and
mortality. This in turn might help in the design of new strategies
for diagnosis and treatment of the disease, thereby allowing for an
improvement in survival rate.
Recently, there has been a
growing interest in investigating various molecular markers in oesophageal carcinoma as potential prognostic,
diagnostic, and perhaps therapeutic tools. "Molecular histology",
a term introduced by Edelman and Crossin in 1991,
characterises the morphological features of a
tissue in terms of the molecules present and the functional
interactions between them. Although cancer invasion and metastasis
form an intricate process, the initial step is the detachment of neoplastic cells from the primary tumour, followed by entry and exit of the lymphatic or
vascular systems and, finally, growth at distant tissue sites. This metastatic cascade of events stems from the dysregulation of normal cellcell adhesion and
cellmatrix interactions. Such interactions are mediated by at least
five families of cell adhesion molecules (CAMs),
namely: integrins, immunoglobulins
(IgCAMs), CD44, selectins,
and cadherins. Apart from regulating
cellcell and cellmatrix interactions, CAMs
also influence cell motility, migration, signalling,
and differentiation, in addition to apoptosis and gene
transcription. In our present article, we provide an overview of
selected CAMs and highlight recent data on
the role of their expression in the regulation of invasion and
metastasis in oesophageal squamous
cell carcinoma (OSCC) and adenocarcinoma (OAC),
and its prognostic relevance.
Journal of Clinical Pathology 58:343-351,2005
Relationship
between HIV viral load and Langerhans cells of the
cervical epithelium
Levi G, Feldman J, Holman S et al
Aim: To determine the relationship between the density
of cervical mucosa Langerhans cells, cervical
histology, and HIV viral load. Methods: Eighty-four HIV-infected and 17 women
at high risk for HIV had cervical biopsies assessed for squamous
intraepithelial lesions and Langerhans cell density. Langerhans cells were identified using the S-100 immunohistochemical stain and were counted manually.
Polymerase chain reaction assays were used to detect cervical human papillomavirus (HPV)-DNA. T-cell subsets were determined
using immunofluorescent flow cytometry.
Plasma HIV RNA levels were measured using a nucleic acid sequence-based
amplification technique. The associations between cervical Langerhans
cell density, cervical histology, CD4 counts, HIV viral loads, HPV-DNA
detection, and smoking status were assessed using multivariate statistical
models. Results: In multivariate analysis among women infected with HIV, the
mean Langerhans cell density per high-powered field
was 4.00 among women with no detectable plasma HIV-RNA, and 1.92 among those
with detectable HIV-RNA (P = 0.01). The mean cervical Langerhans
cell density was increased in women with high-grade squamous
intraepithelial lesions compared with those with low-grade squamous
intraepithelial lesions and normal/metaplastic
histology (3.87 vs 2.11; P = 0.05). Neither HPV-DNA detection, smoking status, nor CD4 count was
significantly associated with Langerhans cell density.
Conclusions: The decrease in cervical Langerhans cell
density in women with detectable HIV-RNA suggests an impaired mucosal immune
response to local infections, such as HPV. Conversely, HPV infection resulting
in high-grade dysplasia might be associated with an
enhanced local immune response.
J Obstet
Gynaecol Res. 31(2):178-84, 2005
CLINICAL PATHOLOGY
Leukocyte Count
as a Predictor of Cardiovascular Events and Mortality in Postmenopausal Women
Karen L. Margolis; JoAnn E. Manson; Philip
Greenland et al
Background Increasing evidence supports a
role for inflammation in the atherosclerotic process. The role of
the leukocyte count as an independent predictor of risk of a first
cardiovascular disease (CVD) event remains uncertain. Our objective
was to describe the relation between the baseline white blood cell
(WBC) count and future CVD events and mortality in postmenopausal women.
Methods In this
prospective cohort study set in 40 US clinical centers, the study
population comprised 72 242 postmenopausal women aged 50 to 79
years, free of CVD and cancer at baseline, enrolled in the Womens
Health Initiative Observational Study. Main outcome measures
included incident fatal coronary heart disease (CHD), nonfatal
myocardial infarction, stroke, and total mortality.
Results At baseline, the
mean SD age of the women was 63 7.3 years,
84% were white, 4% had diabetes, 35% had hypertension, and 6% were
current smokers. The mean WBC count was 5.8 1.6x109
cells/L. During a mean of 6.1 years of follow-up, there were
187 CHD deaths, 701 nonfatal myocardial infarctions, 738 strokes,
and 1919 deaths from all causes. Compared with women with WBC counts
in the first quartile (2.5-4.7x109 cells/L), women in the
fourth quartile (6.7-15.0x109 cells/L) had over a 2-fold
elevated risk for CHD death (hazard ratio, 2.36; 95% confidence
interval, 1.51-3.68), after multivariable adjustment for age, race,
diabetes, hypertension, smoking, hypercholesterolemia, body mass
index, alcohol intake, diet, physical activity, aspirin use, and
hormone use. Women in the upper quartile of the WBC count also had a
40% higher risk for nonfatal myocardial infarction, a 46% higher risk
for stroke, and a 50% higher risk for total mortality. In
multivariable models adjusting for C-reactive protein, the WBC count
was an independent predictor of CHD risk, comparable in magnitude to
C-reactive protein.
Conclusions The WBC count, a stable,
well-standardized, widely available and inexpensive measure of
systemic inflammation, is an independent predictor of CVD events and
all-cause mortality in postmenopausal women. A WBC count greater than 6.7x109 cells/L may
identify high-risk individuals who are not currently identified by
traditional CVD risk factors.
Arch Intern Med. 165:500-508, 2005
MICROBIOLOGY
A
Report of Seven Cases
J. Elliot Carter and Tara N. Evans
To determine the clinical significance of Kluyvera
isolates at their institution, authors retrospectively analyzed clinical
microbiology data from January 1999 to September 2003. Authors identified 11
isolates classified as Kluyvera ascorbata,
7 of which were considered clinically significant pathogens: 3 cases
represented urinary tract infections; 2, bacteremia;
1, a soft tissue infection of the finger; and 1, acute appendicitis with a
subsequent intra-abdominal abscess. The agedistribution
of patients was wide, ranging from 2 months to 73 years. Antimicrobial
susceptibility studies of the clinically significant and nonclinically
significant Kluyvera isolates showed susceptibility
patterns similar to those reported in the medical literature, namely trends of
resistance to ampicillin and first- and
second-generation cephalosporins. Of the 4
nonclinically significant isolates in our study, 1 was resistant to ciprofloxacin, a finding reported in only 1 other isolate of
Kluyvera in the medical literature. Patient outcome
after treatment with third-generation cephalosporins
and aminoglycosides in the 7 clinically significant
cases was good, with no long-term sequelae. The
potential virulence of K ascorbata highlights the
need for heightened scrutiny of its antimicrobial susceptibility patterns for
adequate clinical treatment.
Initially described by Kluyver and van Niel1 in
1936 and documented further by Asai et al in
1956, the bacterial genus Kluyvera was
not defined completely until molecular characterization by Farmer et al in
1981. Previously known as enteric group 8 and also
as API group 1, the genus currently consists of 4 species, Kluyvera
ascorbata, Kluyvera cryocrescens, Kluyvera georgiana (formerly species group 3), and Kluyvera cochleae. Each of these species has
been recovered from human clinical specimens except K cochleae, which has
been isolated from snails and slugs.4 Kluyvera species are present ubiquitously in the environment in water and soil
and also have been described as normal flora of the gastrointestinal tract. 5 The potential virulence
of Kluyvera species has been the subject
of uncertainty in the past, owing in part to its relatively recent
characterization
Am J Clin Pathol 123:334-338, 2005
Arjun
Srinivasan, Elizabeth C. Burton, Matthew J. Kuehnert,
et al
Transplant Recipients Investigation Team
Background In 2004, four
recipients of kidneys, a liver, and an arterial segment from a
common organ donor died of encephalitis of an unknown cause.
Methods: Authors reviewed
the medical records of the organ donor and the recipients. Blood,
cerebrospinal fluid, and tissues from the recipients were tested
with a variety of assays and pathological stains for numerous causes
of encephalitis. Samples from the recipients were also inoculated
into mice.
Results The organ donor
had been healthy before having a subarachnoid hemorrhage
that led to his death. Encephalitis developed in all four recipients
within 30 days after transplantation and was accompanied by rapid neurologic deterioration characterized by
agitated delirium, seizures, respiratory failure, and coma. They
died an average of 13 days after the onset of neurologic
symptoms. Mice inoculated with samples from the affected patients
became ill seven to eight days later, and electron microscopy of
central nervous system (CNS) tissue demonstrated rhabdovirus
particles. Rabies-specific immunohistochemical
and direct fluorescence antibody staining demonstrated rabies virus
in multiple tissues from all recipients. Cytoplasmic
inclusions consistent with Negri bodies
were seen in CNS tissue from all recipients. Antibodies against
rabies virus were present in three of the four recipients and the
donor. The donor had told others of being bitten by a bat.
Conclusions This report
documenting the transmission of rabies virus from an organ donor to
multiple recipients underscores the challenges of preventing and
detecting transmission of unusual pathogens through transplantation.
N Engl J Med 352 (11):1103-1111, 2005
The
interpretation of nucleic acid amplification tests for tuberculosis: do rapid
tests change treatment decisions?
Conaty SJ, Claxton AP, Enoch DA et al
Objectives. To describe changes in treatment decisions
after receipt of nucleic acid amplification (NAA) test for the diagnosis of M.
tuberculosis. Methods. Retrospective notes review of
treatment decisions in patients receiving a NAA test for suspected pulmonary or
non-pulmonary tuberculosis at the
J
Infect. 50(3):187-92, 2005
Newly Discovered
Virus Linked to Childhood Lung Disorders and
A newly discovered virus may be
responsible for many respiratory tract illnesses in infants and children, and
may be associated with an important multi-organ disease whose cause has
remained a mystery for decades, according to articles in the
The virus is one of the numerous coronaviruses, most of which infect animals. In humans, coronaviruses have been known primarily for causing colds
or, more recently, severe acute respiratory syndrome (SARS).
Genetic evidence now suggests
that a previously unknown coronavirus may account for
some of the many respiratory diseases for which a causative agent is
unidentified, and may have a role in
In the first of two studies,
Jeffrey S. Kahn and co-workers at
The Yale investigators, terming
the novel virus indicated by their findings the
That
In an accompanying editorial, Kenneth McIntosh of
The statistically strong
association with