Ronald A. DeLellis

Microvessel quantitation and prognosis in invasive breast carcinoma.

The prognostic significance of microvessel quantitation in invasive breast carcinoma was analyzed in a study group that comprised 88 patients with axillary node-negative carcinoma and 32 patients with axillary node-positive carcinoma who had a minimum follow-up period of 9 years. Microvessels were identified by immunohistochemistry using antibodies to endothelial markers, including factor VIII-related antigen and blood group isoantigens (ABH). Factor VIII-related antigen staining provided more consistent results for microvessel quantitation than did staining for ABH isoantigens. The three most vascular areas within a tumor were selected, and the microvessels within a x200 microscopic field of each area were counted by two investigators simultaneously. Node-positive carcinomas demonstrated significantly higher microvessel counts than did node-negative carcinomas (mean +/- SD, 99 +/- 42 and 73 +/- 22, respectively; P less than .001). In node-negative carcinomas, tumors from patients who experienced distant recurrence had higher microvessel counts than did tumors from patients who were disease-free (84 +/- 19 and 70 +/- 22; P = .01). Similarly, in patients with node-positive carcinoma, microvessel counts were considerably higher in tumors from patients who experienced distant recurrence than in patients who did not, although the difference did not reach statistical significance (113 +/- 44 and 93 +/- 34, respectively). Among patients with node-negative carcinoma, those with a microvessel count of less than 84 had a recurrence rate of 20% compared with 57% in patients with counts greater than 84 (P = .003). Microvessel counts were independent of histologic parameters, ploidy status, and S-phase fraction but correlated with peritumoral vascular invasion. Both microvessel counts and vascular invasion were independent prognostic parameters by multivariate analysis. High vessel counts may represent increased tumor angiogenesis and are correlated with tumor aggressiveness. Microvessel quantitation may be an additional prognostic factor that, when used in conjunction with more established parameters, can help in appropriate patient management.

Observer variation in the diagnosis of follicular variant of papillary thyroid carcinoma

The histopathologic diagnosis of follicular variant of papillary thyroid carcinoma (FVPCA) can be difficult. Recent reports have suggested that this neoplasm may be frequently overdiagnosed by pathologists. We examined the observer variation in the diagnosis of FVPCA in 87 tumors by 10 experienced thyroid pathologists. The criteria that the reviewers considered most helpful for making a diagnosis of FVPCA were also assessed. A concordant diagnosis of FVPCA was made by all 10 reviewers with a cumulative frequency of 39%. In this series, 24.1% of the patients had metastatic disease (n = 21). In the cases with metastatic disease, a diagnosis of FVPCA was made by all 10 reviewers with a cumulative frequency of 66.7%, and 7 of the reviewers made a diagnosis of FVPCA with a cumulative frequency of 100%. The most important criteria used to diagnose FVPCA included the presence of cytoplasmic invaginations into the nucleus (pseudo-inclusions), abundant nuclear grooves, and ground glass nuclei. These results suggest that although the diagnosis of FVPCA is variable even among experienced thyroid pathologists, most reviewers agreed on this diagnosis for patients with metastatic disease. The use of well-defined histopathologic features should improve the consistency in diagnosing FVPCA. Since most cases with metastatic disease had obvious invasion, caution should be used in making a diagnosis of FVPCA in the absence of the major histopathologic features or clear-cut invasive growth.

Distribution of GPR30, a Seven Membrane–Spanning Estrogen Receptor, in Primary Breast Cancer and its Association with Clinicopathologic Determinants of Tumor Progression

Purpose: The seven transmembrane receptor, GPR30, is linked to estrogen binding and heparan-bound epidermal growth factor release. Here, the significance of GPR30 in human breast cancer was evaluated by comparing its relationship to steroid hormone receptor expression and tumor progression variables. Experimental Design: Immunohistochemical analysis of a National Cancer Institute–sponsored tumor collection comprised of 361 breast carcinomas obtained at first diagnosis (321 invasive and 40 intraductal tumors). Biopsies from 12 reduction mammoplasties served as controls. The distribution pattern of GPR30, estrogen receptor (ER), and progesterone receptor (PR) was correlated with clinicopathologic variables obtained at diagnosis. Results: GPR30, ER, and PR were positive in all 12 normal controls. In contrast, GPR30 expression varied in breast tumors, in which 62% (199 of 321) of invasive tumors and 42% (17 of 40) of intraductal tumors were positive. Codistribution of ER and GPR30 was measured in 43% (139 of 321) of invasive breast tumors, whereas both receptors were lacking (ER−GPR30−) in 19% (61 of 321) of the tumors analyzed, indicating a significant association between ER and GPR30 (P < 0.05). The coexpression of PR and ER did not influence GPR30 expression, yet coexpression of GPR30 and ER was linked to PR positivity. Unlike ER, which varied inversely with HER-2/neu and tumor size, GPR30 positively associated with HER-2/neu and tumor size. In addition, GPR30 showed a positive association with metastasis (P = 0.014; odds ratio, 1.9). Conclusions: GPR30 and ER exhibited distinct patterns of association with breast tumor progression variables, including HER-2/neu, tumor size, and metastatic disease. Thus, these results support the hypothesis that GPR30 and ER have an independent influence on estrogen responsiveness in breast carcinoma.

Reversibility of hepatic fibrosis in autoimmune hepatitis.

Hepatic fibrosis occurs in response to many types of chronic hepatic injury. When fibrosis is extensive and distorts the normal hepatic architecture with regenerative nodules, it is called cirrhosis. Fibrosis and cirrhosis are believed to be irreversible [1, 2]. However, anecdotal reports have suggested their reversibility in hemochromatosis [3] and Wilson disease [4], and we recently documented their regression in several patients with primary biliary cirrhosis [5]. We describe eight patients with autoimmune, steroid-responsive hepatitis in whom cirrhosis or extensive fibrosis was confirmed on liver biopsy at diagnosis but disappeared or was markedly reduced in response to effective treatment. Methods Study Sample Eight patients were retrospectively selected because they had extensive scarring or cirrhosis on liver biopsy or laparotomy at time of diagnosis; responded to treatment with glucocorticoids, immunosuppressive drugs, or both; and had liver biopsy some years after treatment began. They were part of a group of 97 patients with a diagnosis of autoimmune hepatitis who were seen at New England Medical Center between 1966 and 1996. The other 89 patients did not have cirrhosis at diagnosis (n = 55), had cirrhosis at diagnosis but did not have liver biopsy after successful medical treatment (n = 25), had cirrhosis at diagnosis and still had extensive fibrosis on liver biopsy despite symptomatic and biochemical response to treatment (n = 6), or did not have a liver biopsy specimen of adequate size (>5 mm) at diagnosis or follow-up (n = 3). The diagnosis of autoimmune hepatitis was based primarily on a complete response to steroids [6], increased aminotransferase and total globulin levels, chronic active hepatitis on histologic study of the liver, and exclusion of other causes of hepatitis (such as alcohol or viral infection) [7]. Patients given a diagnosis before 1989 were not tested for hepatitis C virus. Histologic Assessment All liver biopsy specimens were measured; those 5 mm or less in length were excluded. All slides were coded in an unpaired manner and read blindly and independently by a hepatologist and a liver pathologist using the Knodell scoring system [8]. Case Reports Patient 1 A 40-year-old man had surgical cholecystectomy and wedge liver biopsy in 1970 because of fever and abnormal results on liver function tests. His aspartate aminotransferase (AST) level was 7.50 kat/L, and his alanine aminotransferase (ALT) level was 4.33 kat/L. The liver was nodular. Liver biopsy showed portal and lobular inflammation with extensive fibrosis (Figure 1, part A). The patient was treated with prednisone, 30 mg/d. His ALT and AST levels returned to normal within 6 months. Prednisone therapy was tapered and discontinued after 2 years. Blood test results remained normal for 8 years, after which time autoimmune hepatitis recurred. The patient noted fatigue and fever, and his ALT and AST levels were elevated sevenfold. Response to prednisone was again rapid. Liver biopsy done 2 years later had normal results with the exception of minimal portal fibrosis (Figure 1, part B). The patient had additional relapses while not receiving treatment 12 and 20 years after diagnosis. Each relapse responded quickly to prednisone therapy. An attempt to use azathioprine for maintenance therapy was unsuccessful because of an idiosyncratic reaction. Figure 1. Liver biopsy specimens. Patient 2 A 17-year-old man presented with jaundice, an AST level of 20.25 kat/L, an ALT level of 22.09 kat/L, and hyperglobulinemia (globulin level, 62 g/L). He received prednisone, 20 mg/d, and azathioprine, 100 mg/d. After 6 months, his ALT level was 0.98 kat/L and his AST level was 1.20 kat/L. The patient discontinued treatment against medical advice and was untreated for 2 years. His ALT level increased to 19.09 kat/L, and his AST level increased to 25.42 kat/L. Liver biopsy showed chronic hepatitis and cirrhosis (Figure 1, part C). The patient resumed treatment with prednisone, 30 mg/d, and azathioprine, 100 mg/d. His ALT level decreased to 0.27 kat/L, and his AST level decreased to 0.45 kat/L. Prednisone therapy was discontinued after 2 years, and the patient remained in remission with azathioprine alone. Liver biopsy done 8 years after diagnosis showed no sign of hepatitis and minimal fibrosis (Figure 1, part D). Patient 3 A 34-year-old woman had been told that her liver function test results were abnormal 17 years earlier. Her ALT level was 8.30 kat/L at time of referral. Liver biopsy showed chronic active hepatitis and cirrhosis (Figure 1, part E). The patient began receiving prednisone, 20 mg/d, and her aminotransferase levels returned to normal. Attempts to taper steroid dose and maintain the patient with azathioprine or 6-mercaptopurine were unsuccessful because of adverse reactions. After 2 years of prednisone therapy, the ALT level was 0.28 kat/L and the AST level was 0.33 kat/L. Repeated liver biopsy showed nearly normal results with minimal portal inflammation and marked regression of fibrosis (Figure 1, part F). The patient remained well while receiving prednisone for 8 years. Results The final study group included six male and two female patients. The median age at presentation was 36 years (range, 8 to 61 years). At presentation, all patients had elevated ALT levels (median, 10.30 kat/L [range, 4.33 to 36.26 kat/L]), AST levels (median, 12.79 kat/L [range, 5.15 to 35.42 kat/L]), and globulin levels (median, 43 g/L [range, 31 to 62 g/L]) (Table 1). The median albumin level was 34 g/L (range, 31 to 42 g/L), and the median platelet count was 143 000 cells/mm3 (range, 103 000 cells/mm3 to 200 000 cells/mm3). All patients received steroids; four also received 6-mercaptopurine and four also received azathioprine. At the time of the final biopsy, ALT levels were significantly lower than at presentation (median, 0.37 kat/L [range, 0.17 to 1.50 kat/L]), as were AST levels (median, 0.35 kat/L [range, 1.33 to 0.25 kat/L]) and globulin levels (median, 25 U/L [range, 20 to 34 U/L]). Serum albumin levels increased during therapy (median, 43 g/L [range, 39 to 48 g/L]), as did platelet counts (median, 200 000 cells/mm3 [range, 110 000 to 364 cells/mm3]) (Table 1). The median duration of follow-up was 9.5 years (range, 4 to 28 years). After follow-up, four patients were in remission while receiving no treatment, two were receiving 6-mercaptopurine, and two were receiving prednisone. Table 1. Eight Patients with Autoimmune Hepatitis before and after Successful Medical Treatment* The median length of biopsy specimens at presentation was 15 mm (range, 13 to 25 mm for seven of the eight specimens; one specimen was from a wedge biopsy). The median length of final biopsy specimens was 21 mm (range, 6 mm to 32 mm for all eight specimens). The median Knodell score at presentation was 14.0, with a median fibrosis grade of 3.3 (Table 1). The median Knodell score on the final set of biopsy specimens was 1.3, with a median fibrosis grade of 0.8 (Table 1). The median interval between initial and final biopsy was 47 months (range, 13 to 118 months). There was good concordance between the pathologists, as shown in the (Table 1). Discussion We describe eight patients who had documented cirrhosis or extensive fibrosis at the time of diagnosis of autoimmune hepatitis. Blood test results became normal in response to medical therapy, and subsequent biopsies showed complete regression of hepatic fibrosis. Several trials have reported the successful long-term treatment of autoimmune hepatitis and commented on the induction of clinical, biochemical, and histologic remission. None, to the best of our knowledge, has discussed the reversibility of fibrosis. Fibrosis results from the accumulation of connective tissue and implies an imbalance between the deposition and the degradation of that tissue. The pathogenesis of progressive scarring seems similar for different types of hepatic injuries. Schematically, lipocytes activated by cytokines, such as transforming growth factor 1, degrade the normal extracellular matrix and replace it with fibril-forming collagens types I and III [9]. Less is known about fibrolysis. The liver contains metalloproteinases and collagenases capable of degrading extracellular matrix. Collagen degradation is a slow process, however, and collagen I, after its deposition, sustains extensive cross-linking and becomes more resistant to collagenases over time [10]. Regression of hepatic fibrosis has been seen in animal models. In rats, functional and structural recovery occurred with partial reversibility of fibrosis 1 month after decompression of chronic bile duct obstruction [11]. In rabbits, liver fibrosis due to Schistosoma japonicum infection regressed after parasitologic cure [12]. In humans, liver fibrosis and cirrhosis are generally believed to be irreversible. At the same time, anecdotal reports suggest improvement of cirrhosis in patients treated for hemochromatosis [3] and Wilson disease [4]; we recently documented regression of fibrosis in several patients with primary biliary cirrhosis [5]. In autoimmune hepatitis, effective therapy rapidly controls hepatic inflammation and hepatocellular necrosis. Treatment may decrease the inciting stimulus for fibrogenesis and unmask or upregulate the fibrolytic side of the dynamic matrix remodelling process. However, cirrhosis seems to be reversible only in patients who are in the early clinical stages of disease. Our eight patients were Child-Pugh class A or B. Cirrhosis is probably irreversible in patients with advanced liver disease (Child-Pugh class C with portal hypertension). Our study has some limitations. This was a small group of selected patients studied retrospectively, and sampling artifacts of the liver biopsy specimens (despite the generous size of the specimens) may account for the differences in fibrosis [13]. To minimize bias, liver biopsy specimens were read under code and were not paired

Observer variation of encapsulated follicular lesions of the thyroid gland

Although histologic definition of follicular thyroid lesions is readily available, application of the diagnostic criteria and personal experience may lead to disagreement among pathologists. To investigate interobserver variation in assessment of encapsulated follicular lesions, eight pathologists (four American and four Japanese) reviewed the same hematoxylin and eosin-stained slide of each of 21 cases of thyroid lesions showing encapsulation and follicular growth pattern. In 10% of the cases, there was complete agreement. At least seven pathologists agreed on the diagnosis in 29% of the cases, and at least six in 76% of the cases. American and Japanese pathologists agreed among themselves in 33% and 52% of cases, respectively. The frequency of diagnosis of adenomatous goiter among Japanese pathologists (31%) was considerably higher than that among American pathologists (6%). In contrast, the frequency of diagnosis (25%) of papillary carcinoma among American pathologists was considerably higher than that (4%) among Japanese pathologists. Our analysis revealed three main factors affecting observer variation: 1) interpretation of the significance of microfollicles intimately related to capillaries within the tumor capsule, 2) evaluation of what constituted the type of nuclear clearing indicative of papillary carcinoma, and 3) absence of clear morphologic criteria for separation of adenomatous goiter and follicular adenoma. To reduce observer variation of encapsulated follicular lesions, it will be necessary to provide more explicit criteria for diagnosis.

Interobserver and Intraobserver Variation Among Experts in the Diagnosis of Thyroid Follicular Lesions With Borderline Nuclear Features of Papillary Carcinoma

Distinguishing follicular variant of papillary carcinoma (FVPC) from follicular adenoma and follicular carcinoma can be difficult if nuclear features of papillary carcinoma are not well developed or only focally present. We assessed interobserver and intraobserver agreement among 6 thyroid experts by using 15 cases in which original pathologists suspected FVPC. There was unanimous expert agreement in diagnosing FVPC in only 2 cases (13%) and majority agreement in 6 cases (40%). Unanimous agreement on benign and malignant diagnoses was seen in 4 cases (27%) and majority agreement on malignancy in 8 cases (53%). Intraobserver agreement ranged from 17% to 100%. Histologic features considered most helpful in diagnosing FVPC were nuclear clearing, nuclear grooves, nuclear overlapping and crowding, nuclear membrane irregularity, and nuclear enlargement. This considerable interobserver and intraobserver variability in the diagnosis of FVPC seems to result from lack of agreement on the minimal criteria needed to diagnose FVPC, even among experts.

Adenoma malignum (minimal deviation adenocarcinoma) of the uterine cervix. A clinicopathological and immunohistochemical analysis of 26 cases

We reviewed 26 examples of the rare variant of cervical adenocarcinoma that has been designated “adenoma malignum.” The patients, three of whom had Peutz-Jeghers syndrome, ranged in age from 25 to 72 years (average, 42 years). The most common presenting symptom was menometrorrhagia, followed by vaginal discharge, postmenopausal bleeding, and abdominal swelling in decreasing order of frequency. In 12 of the patients, the diagnosis was established on the basis of the examination of a cervical biopsy specimen, endocervical curettage specimen, or both. In three of these cases, however, up to four biopsies were performed before the diagnosis was established. In the remaining 14 patients, the diagnosis was not made until the time of operation or pathologic examination of a hysterectomy specimen. On gross examination, the cervix usually appeared abnormal, but occasional specimens were considered unremarkable. The cervix was typically described as firm or indurated. Microscopic examination showed glands that were irregular in size and shape and lined predominantly by mucin-containing columnar epithelial cells with basal nuclei. The tumors typically exhibited deep invasion of the cervical wall, and a portion of the infiltrating tumor was associated with a stromal response in most cases. Minor foci of tumor with a less well-differentiated appearance were present in 15 of the 26 tumors. Argyrophil cells were present in six of 15 tumors. Five of the six tumors containing argyrophil cells stained immunohistochemically for serotonin and peptide hormones. Positive staining for serotonin was seen in four tumors; one of these also contained a few cells positive for neurotensin. Cytoplasmic staining of the tumor cells for carcinoembryonic antigen (CEA) was seen in five of six cases. CEA reactivity was very focal in two of the positive tumors. Microsocopic features that were most helpful in distinguishing adenoma malignum from normal endocervix or benign endocervical glandular proliferations were the presence of markedly irregular, abnormally shaped glands; invasion of the cervical wall; a loose edematous or desmoplastic stromal response; foci of less well-differentiated tumor; vascular invasion; perineural invasion; and positive staining for CEA. Despite radical therapy in most of the cases, the prognosis was poor. Follow-up data were available for 22 patients. Thirteen of them died of recurrent tumor, four were alive with recurrent tumor at the time of last follow-up examination, and only three patients were disease free for 2 years or more. Four of 10 patients who had at least one ovary conserved at the time of initial operation subsequently had a mucinous tumor of the ovary, indicating a risk of ovarian conservation in these patients.

Primary extranodal lymphoma: response to treatment and factors influencing prognosis.

The cumulative 10 year lymphoma experience of a teaching hospital and two of its affiliated institutions was reviewed. From this group, a series of 39 cases of regionally localized primary extranodal lymphomas (Ann Arbor Stages IE and IIE) were selected for study. This group of patients was analyzed for response to initial curative treatment and factors influencing prognosis. The disease‐free survival rate following initial treatment is 41% and the rate is 51% if those treated for a single relapse are included. Factors which clearly influence prognosis in this group are stage (extent of disease) at presentation and Rappaport histologic subclassification. The relationships of anatomic site and age to prognosis independent of other factors are unclear. There appears to be an association between sites of involvement in Waldeyers Ring and the gastrointestinal tract seen both initially and in sites of relapse. An analysis of relapse patterns revealed that 33% of relapses occur as solitary extranodal “skip” recurrences that when treated with radical local treatment may result in long disease‐free survival. The latter is one of several findings which suggest that a modification of the current Ann Arbor staging system may be necessary to encompass certain unique features of this group of tumors.

Natural history of familial medullary thyroid carcinoma: effect of a program for early diagnosis.

To detect familial medullary thyroid carcinoma in a premetastatic stage, we administered tests provocative of calcitonin secretion (infusion of calcium or pentagastrin or both) each year for seven years to members of a pedigree now numbering 107. Since 1970, 21 patients converted from normal to abnormal secretory responses (two separate tests in which calcitonin levels exceeded 0.58 ng per milliliter). Twenty of 21 glands removed showed C-cell hyperplasia, and eight of the 20 also showed foci of carcinoma. As compared to the 12 patients with tumors detected during the first year of screening, all of whom had bilateral carcinoma (seven of 12 with local metastases), later carcinomas were smaller (mean diameter of 0.2 vs. 0.8 cm), were unilateral (in all but two cases) and occurred in younger patients (mean age of 14.9 vs. 36.4 years), and none had detectable metastases.

Distribution of Calcitonin-Containing Cells in the Normal Neonatal Human Thyroid Gland: A Correlation of Morphology with Peptide Content

C-cells have been mapped in the thyroid glands of 6 human neonates by means of immunoperoxidase localization of calcitonin and tissue calcitonin content as measured by radioimmunoassay. The C-cells were concentrated in a zone in the upper two-thirds of the lateral lobes bilaterally, where they were identified individually and in small groups in both an intrafollicular and parafollicular distribution. In contrast to findings in the adult, C-cells were predominantly intrafollicular in the neonate. The relative numbers of C-cells counted per unit area of thyroid tissue correlated strongly with the calcitonin content of immediately adjacent tissue sections. In areas rich in C-cells, as many as 75 immunoperoxidase-stained cells per low-power field were counted, and the concentration of calcitonin was as high as 540 to 2100 mU/g fresh weight, values that were as great as 10 times those observed in the normal adult thyroid gland. The prominence of the C-cell population and increased tissue calcitonin content in the human neonatal thyroid gland may reflect an as yet undefined physiologic role for calcitonin in the newborn.