Barriers and facilitators of artificial intelligence conception and implementation for breast imaging diagnosis in clinical practice: a scoping review.

OBJECTIVE: Although artificial intelligence (AI) has demonstrated promise in enhancing breast cancer diagnosis, the implementation of AI algorithms in clinical practice encounters various barriers. This scoping review aims to identify these barriers and facilitators to highlight key considerations for developing and implementing AI solutions in breast cancer imaging. METHOD: A literature search was conducted from 2012 to 2022 in six databases (PubMed, Web of Science, CINHAL, Embase, IEEE, and ArXiv). The articles were included if some barriers and/or facilitators in the conception or implementation of AI in breast clinical imaging were described. We excluded research only focusing on performance, or with data not acquired in a clinical radiology setup and not involving real patients. RESULTS: A total of 107 articles were included. We identified six major barriers related to data (B1), black box and trust (B2), algorithms and conception (B3), evaluation and validation (B4), legal, ethical, and economic issues (B5), and education (B6), and five major facilitators covering data (F1), clinical impact (F2), algorithms and conception (F3), evaluation and validation (F4), and education (F5). CONCLUSION: This scoping review highlighted the need to carefully design, deploy, and evaluate AI solutions in clinical practice, involving all stakeholders to yield improvement in healthcare. CLINICAL RELEVANCE STATEMENT: The identification of barriers and facilitators with suggested solutions can guide and inform future research, and stakeholders to improve the design and implementation of AI for breast cancer detection in clinical practice. KEY POINTS: • Six major identified barriers were related to data; black-box and trust; algorithms and conception; evaluation and validation; legal, ethical, and economic issues; and education. • Five major identified facilitators were related to data, clinical impact, algorithms and conception, evaluation and validation, and education. • Coordinated implication of all stakeholders is required to improve breast cancer diagnosis with AI.

ESUR recommendations for MR imaging of the sonographically indeterminate adnexal mass: an update.

An update of the 2010 published ESUR recommendations of MRI of the sonographically indeterminate adnexal mass integrating functional techniques is provided. An algorithmic approach using sagittal T2 and a set of transaxial T1 and T2WI allows categorization of adnexal masses in one of the following three types according to its predominant signal characteristics. T1 'bright' masses due to fat or blood content can be simply and effectively determined using a combination of T1W, T2W and FST1W imaging. When there is concern for a solid component within such a mass, it requires additional assessment as for a complex cystic or cystic-solid mass. For low T2 solid adnexal masses, DWI is now recommended. Such masses with low DWI signal on high b value image (e.g. > b 1000 s/mm2) can be regarded as benign. Any other solid adnexal mass, displaying intermediate or high DWI signal, requires further assessment by contrast-enhanced (CE)T1W imaging, ideally with DCE MR, where a type 3 curve is highly predictive of malignancy. For complex cystic or cystic-solid masses, both DWI and CET1W-preferably DCE MRI-is recommended. Characteristic enhancement curves of solid components can discriminate between lesions that are highly likely malignant and highly likely benign. KEY POINTS: • MRI is a useful complementary imaging technique for assessing sonographically indeterminate masses. • Categorization allows confident diagnosis in the majority of adnexal masses. • Type 3 contrast enhancement curve is a strong indicator of malignancy. • In sonographically indeterminate masses, complementary MRI assists in triaging patient management.

Are Second Trimester Apparent Diffusion Coefficient Values of the Short Uterine Cervix Associated with Impending Preterm Delivery?

Purpose To prospectively determine whether the apparent diffusion coefficient (ADC) of the cervix is associated with preterm delivery in asymptomatic patients with a sonographic cervical length of 15 mm or less and positive fetal fibronectin test results between 23 and 28 weeks of gestation. Materials and Methods The institutional review board approved this prospective hypotheses-generating study. A total of 30 pregnant women (mean gestational age, 26 weeks) with a sonographic short cervix (≤15 mm) underwent pelvic 1.5-T magnetic resonance (MR) imaging. Oblique sagittal diffusion-weighted images were obtained with b values of 0, 400, and 800 sec/mm(2). ADC values at MR imaging of the subglandular and stromal cervix and the difference between both were correlated to the interval to delivery. Receiver operating characteristic curve analysis was performed to obtain sensitivity and specificity of ADC values in association with delivery within 7 days. Results Eight (27%) of 30 patients delivered within 6 or 7 days after MR imaging (impending delivery group), and 22 (73%) of 30 patients delivered between 18 and 89 days after imaging (mean, 55 days) (late delivery group). Mean subglandular ADC and mean ADC difference were higher (P < .001) in patients with impending delivery than in those with late delivery ([2406.3 ± 166.0] × 10(-6) mm(2)/sec vs [1708.9 ± 108.1] × 10(-6) mm(2)/sec and [657.3 ± 129.9] × 10(-6) mm(2)/sec vs [69.2 ± 70.2] × 10(-6) mm(2)/sec, respectively). Subglandular ADC inversely correlated with the interval between MR imaging and delivery (r = -0.75). Receiver operating characteristic curve analysis of subglandular ADC revealed 100% sensitivity (95% confidence interval: 63.1, 100) and 100% specificity (95% confidence interval: 84.6, 100) in association with impending delivery with a 1921 × 10(-6) mm(2)/sec threshold. Stromal ADC and sonographic cervical length showed no difference between groups (P = .072 and P = .511, respectively). Conclusion Cervical subglandular ADC at MR imaging is associated with impending preterm birth in patients with a short sonographic cervix. (©) RSNA, 2016.

Efficient Clinical Information Extraction from Breast Radiology Reports in French.

Radiology reports contain crucial patient information, in addition to images, that can be automatically extracted for secondary uses such as clinical support and research for diagnosis. We tested several classifiers to classify 1,218 breast MRI reports in French from two Swiss clinical centers. Logistic regression performed better for both internal (accuracy > 0.95 and macro-F1 > 0.86) and external data (accuracy > 0.81 and macro-F1 > 0.41). Automating this task will facilitate efficient extraction of targeted clinical parameters and provide a good basis for future annotation processes through automatic pre-annotation.

ESUR guidelines: ovarian cancer staging and follow-up.

OBJECTIVE: To design clear guidelines for the staging and follow-up of patients with ovarian cancer, and to provide the radiologist with a framework for use in multidisciplinary conferences. METHODS: Guidelines for ovarian cancer staging and follow-up were defined by the female imaging subcommittee of the ESUR (European Society of Urogenital Radiology) based on the expert consensus of imaging protocols of 12 leading institutions and a critical review of the literature. RESULTS: Computed tomography (CT) with coverage of the base of the lungs to the inguinal region is regarded as the imaging technique of choice for preoperative staging. Critical diagnostic criteria are presented and the basis for a structured report for preoperative staging is outlined. Following primary treatment for ovarian cancer, clinical assessment and CA-125 are routinely used to monitor patients. For suspected recurrence, CT remains the imaging modality of choice, with positron emission tomography (PET)/CT emerging as the optimal imaging technique for suspected recurrence, particularly in patients with negative CT or magnetic resonance imaging (MRI). CONCLUSIONS: CT is the imaging modality of choice for preoperative staging and detection of recurrence in patients with ovarian cancer.

ESUR guidelines for MR imaging of the sonographically indeterminate adnexal mass: an algorithmic approach.

A significant proportion of adnexal masses detected by sonography are indeterminate. Either their organ of origin is uncertain or it is unclear whether their nature is benign or malignant. MR imaging of the sonographically indeterminate adnexal mass can resolve most of these uncertainties. Most indeterminate masses result from common benign conditions and women with such masses can avoid unnecessary or inappropriate surgery. For the minority of women whose masses are malignant, use of MR imaging rather than a 'wait and watch' strategy of repeat ultrasound (US) results in a more timely diagnosis. There are simple diagnostic steps in the MR imaging assessment which direct an algorithmic and problem-solving approach based on signal characteristics and morphology. MR imaging should provide a more timely diagnosis and, thereby, guide the management of the patient with reduced costs of investigation and treatment.

Image-guided breast biopsy and localisation: recommendations for information to women and referring physicians by the European Society of Breast Imaging.

We summarise here the information to be provided to women and referring physicians about percutaneous breast biopsy and lesion localisation under imaging guidance. After explaining why a preoperative diagnosis with a percutaneous biopsy is preferred to surgical biopsy, we illustrate the criteria used by radiologists for choosing the most appropriate combination of device type for sampling and imaging technique for guidance. Then, we describe the commonly used devices, from fine-needle sampling to tissue biopsy with larger needles, namely core needle biopsy and vacuum-assisted biopsy, and how mammography, digital breast tomosynthesis, ultrasound, or magnetic resonance imaging work for targeting the lesion for sampling or localisation. The differences among the techniques available for localisation (carbon marking, metallic wire, radiotracer injection, radioactive seed, and magnetic seed localisation) are illustrated. Type and rate of possible complications are described and the issue of concomitant antiplatelet or anticoagulant therapy is also addressed. The importance of pathological-radiological correlation is highlighted: when evaluating the results of any needle sampling, the radiologist must check the concordance between the cytology/pathology report of the sample and the radiological appearance of the biopsied lesion. We recommend that special attention is paid to a proper and tactful approach when communicating to the woman the need for tissue sampling as well as the possibility of cancer diagnosis, repeat tissue sampling, and or even surgery when tissue sampling shows a lesion with uncertain malignant potential (also referred to as "high-risk" or B3 lesions). Finally, seven frequently asked questions are answered.

Mammographic texture synthesis: second-generation clustered lumpy backgrounds using a genetic algorithm.

Synthetic yet realistic images are valuable for many applications in visual sciences and medical imaging. Typically, investigators develop algorithms and adjust their parameters to generate images that are visually similar to real images. In this study, we used a genetic algorithm and an objective, statistical similarity measure to optimize a particular texture generation algorithm, the clustered lumpy backgrounds (CLB) technique, and synthesize images mimicking real mammograms textures. We combined this approach with psychophysical experiments involving the judgment of radiologists, who were asked to qualify the visual realism of the images. Both objective and psychophysical approaches show that the optimized versions are significantly more realistic than the previous CLB model. Anatomical structures are well reproduced, and arbitrary large databases of mammographic texture with visual and statistical realism can be generated. Potential applications include detection experiments, where large amounts of statistically traceable yet realistic images are needed.

Breast ultrasound: recommendations for information to women and referring physicians by the European Society of Breast Imaging.

This article summarises the information that should be provided to women and referring physicians about breast ultrasound (US). After explaining the physical principles, technical procedure and safety of US, information is given about its ability to make a correct diagnosis, depending on the setting in which it is applied. The following definite indications for breast US in female subjects are proposed: palpable lump; axillary adenopathy; first diagnostic approach for clinical abnormalities under 40 and in pregnant or lactating women; suspicious abnormalities at mammography or magnetic resonance imaging (MRI); suspicious nipple discharge; recent nipple inversion; skin retraction; breast inflammation; abnormalities in the area of the surgical scar after breast conserving surgery or mastectomy; abnormalities in the presence of breast implants; screening high-risk women, especially when MRI is not performed; loco-regional staging of a known breast cancer, when MRI is not performed; guidance for percutaneous interventions (needle biopsy, pre-surgical localisation, fluid collection drainage); monitoring patients with breast cancer receiving neo-adjuvant therapy, when MRI is not performed. Possible indications such as supplemental screening after mammography for women aged 40-74 with dense breasts are also listed. Moreover, inappropriate indications include screening for breast cancer as a stand-alone alternative to mammography. The structure and organisation of the breast US report and of classification systems such as the BI-RADS and consequent management recommendations are illustrated. Information about additional or new US technologies (colour-Doppler, elastography, and automated whole breast US) is also provided. Finally, five frequently asked questions are answered. TEACHING POINTS: • US is an established tool for suspected cancers at all ages and also the method of choice under 40. • For US-visible suspicious lesions, US-guided biopsy is preferred, even for palpable findings. • High-risk women can be screened with US, especially when MRI cannot be performed. • Supplemental US increases cancer detection but also false positives, biopsy rate and follow-up exams. • Breast US is inappropriate as a stand-alone screening method.

Semiautomatic mammographic parenchymal patterns classification using multiple statistical features.

RATIONALE AND OBJECTIVES: Our project was to investigate a complete methodology for the semiautomatic assessment of digital mammograms according to their density, an indicator known to be correlated to breast cancer risk. The BI-RADS four-grade density scale is usually employed by radiologists for reporting breast density, but it allows for a certain degree of subjective input, and an objective qualification of density has therefore often been reported hard to assess. The goal of this study was to design an objective technique for determining breast BI-RADS density. MATERIALS AND METHODS: The proposed semiautomatic method makes use of complementary pattern recognition techniques to describe manually selected regions of interest (ROIs) in the breast with 36 statistical features. Three different classifiers based on a linear discriminant analysis or Bayesian theories were designed and tested on a database consisting of 1408 ROIs from 88 patients, using a leave-one-ROI-out technique. Classifications in optimal feature subspaces with lower dimensionality and reduction to a two-class problem were studied as well. RESULTS: Comparison with a reference established by the classifications of three radiologists shows excellent performance of the classifiers, even though extremely dense breasts continue to remain more difficult to classify accurately. For the two best classifiers, the exact agreement percentages are 76% and above, and weighted kappa values are 0.78 and 0.83. Furthermore, classification in lower dimensional spaces and two-class problems give excellent results. CONCLUSION: The proposed semiautomatic classifiers method provides an objective and reproducible method for characterizing breast density, especially for the two-class case. It represents a simple and valuable tool that could be used in screening programs, training, education, or for optimizing image processing in diagnostic tasks.

Paget disease of the breast: findings at magnetic resonance imaging and histopathologic correlation.

OBJECTIVES: The purpose of this article is to describe magnetic resonance imaging (MRI) findings in patients with Paget disease of the breast and to evaluate mammography and MRI of the breast in the diagnosis of associated breast cancer. MATERIALS AND METHODS: Nine patients with biopsy-proven Paget disease of the nipple underwent preoperative mammography and MRI of the breast to evaluate underlying breast cancer. All patients underwent subsequent surgery. The patients' charts and imaging studies were retrospectively reviewed. Imaging findings were correlated to results of histopathology. RESULTS: Histopathology confirmed Paget disease of the nipple in all 9 patients and diagnosed associated ductal carcinoma in situ (DCIS) in the retroareolar lactiferous ducts in 8 of 9 patients (88%). MRI showed abnormal nipple enhancement in these 8 patients with an ill-defined thickened nipple-areolar complex. DCIS elsewhere in the breast was diagnosed in 4 of 9 patients (45%) corresponding to nonfocal enhancement in all 4 patients at MRI of the breast (100%). CONCLUSIONS: Paget disease of the breast associated with underlying DCIS can be diagnosed at MRI of the breast and therefore impact management decisions.

Computer-aided detection (CAD) in mammography: does it help the junior or the senior radiologist?

OBJECTIVES: To evaluate the impact of a computer-aided detection (CAD) system on the ability of a junior and senior radiologist to detect breast cancers on mammograms, and to determine the potential of CAD as a teaching tool in mammography. METHODS: Hundred biopsy-proven cancers and 100 normal mammograms were randomly analyzed by a CAD system. The sensitivity (Se) and specificity (Sp) of the CAD system were calculated. In the second phase, to simulate daily practice, 110 mammograms (97 normal or with benign lesions, and 13 cancers) were examined independently by a junior and a senior radiologist, with and without CAD. Interpretations were standardized according to BI-RADS classification. Sensitivity, Specificity, positive and negative predictive values (PPV, NPV) were calculated for each session. RESULTS: For the senior radiologist, Se slightly improved from 76.9 to 84.6% after CAD analysis (NS) (one case of clustered microcalcifications case overlooked by the senior radiologist was detected by CAD). Sp, PPV and PNV did not change significantly. For the junior radiologist, Se improved from 61.9 to 84.6% (significant change). Three cancers overlooked by the junior radiologist were detected by CAD. Sp was unchanged. CONCLUSION: CAD mammography proved more useful for the junior than for the senior radiologist, improving sensitivity. The CAD system may represent a useful educational tool for mammography.

Diagnosis of endometriosis: pelvic endoscopy and imaging techniques.

Although laparoscopy is the gold standard for the diagnosis of endometriosis, the need for this invasive diagnostic tool is a major stumbling-block in both effective clinical management and for research into this common and debilitating disease. As visual inspection of the pelvis has also major limitations, particularly for the diagnosis of posterior pelvis, bowel and bladder endometriosis, it is not surprising that considerable efforts are being made to improve the diagnosis by imaging techniques. Peritoneal endometriosis and ovarian endometriomata are predominantly haemorrhagic lesions. During laparoscopy, these lesions are readily identified by the presence of old or recent bleeding. Both transvaginal sonography and magnetic resonance (MR) imaging have a low sensitivity for the diagnosis of peritoneal and ovarian implants and adhesions. Transvaginal sonography is useful in the diagnosis of ovarian endometriomata, providing the diameter is larger than 2 cm. As transvaginal sonography costs less than MR imaging, transvaginal sonography might be the preferred method of confirming a sizeable endometrioma. Posterior pelvis endometriosis is an infiltrating adenomyotic lesion with microendometriomata. High-resolution transvaginal ultrasonography, and in particular MR imaging, are increasingly used to diagnose the presence and extent of infiltrating lesions and the involvement of rectosigmoid and ureters.

Magnetic resonance imaging of the cervix in women at high risk for preterm delivery.

OBJECTIVES: To assess whether changes in signal intensity of cervical stroma layers on magnetic resonance imaging (MRI) are associated with spontaneous preterm delivery. METHODS: Prospective cohort study of women admitted for threatened late miscarriage or preterm delivery between 18 and 34 weeks of gestation. We performed T2-weighted low-field MRI of the uterine cervix among 100 women. Cervical stromal differentiation, defined as the contrast between signal intensities of the inner and outer cervical layers, was classified as high, intermediate, or low by a radiologist blinded to the participant's clinical report. The main outcome measure was the proportion of spontaneous preterm delivery. RESULTS: Thirty-six women had a spontaneous preterm delivery. The proportion of spontaneous preterm delivery for high, intermediate, and low stromal differentiation was 7/24 (29%), 21/64 (33%; risk ratio 1.1; 95% confidence interval [CI]: 0.6-2.3), and 8/12 (67%; risk ratio 2.3; 95% CI: 1.1-4.8), respectively. The risk of delivering within 7 days increased when stromal differentiation decreased, although the difference was not statistically significant. CONCLUSIONS: The risk of spontaneous preterm delivery is increased in women with low cervical stromal differentiation on MRI. This risk is also associated with short cervical length, a measurement easier and less costly to obtain by transvaginal ultrasound.

Histopathological study of breast cancer and normal breast tissue after magnetic resonance-guided cryotherapy ablation.

BACKGROUND: Cryotherapy ablation is a minimally invasive procedure being investigated as an alternative to conventional surgery. There are few reports in breast cancer. AIM: Evaluate the histopathology of tumoral and normal breast tissue after cryotherapy. METHODS: Eleven patients with clinically <2.0cm and ultrasound visible tumors were studied. Invasive carcinoma was documented by preoperative mammography, magnetic resonance imaging and biopsies. Cryotherapy needles were inserted in the tumor under magnetic resonance guidance and deep freezed with a CRYO-HIT TM System-3. Lumpectomy was performed within 4-5 weeks following cryoablation and submitted for pathological examination including immunostaining of keratins. RESULTS: The tumoral response after cryoablation was variable. In 4 cases there was no viable invasive carcinoma left and focal DCIS only in 2. In 6 cases, residual invasive carcinoma of various size was present with DCIS inside or outside the cryozone. One case could not be evaluated because the cryozone was adjacent to the tumor due to technical problems. Histologically, the normal breast parenchyma of the cryozone showed dense fibrosis, fat necrosis, xanthogranulomatous reaction, endovascular fibrosis and haemorrhages in all cases. The positive immunostaining of keratins revealed remnants of cytoskeleton of carcinomatous cells in the necrotic areas without any viable tumoral cells on routine stains. Skin ulceration and/or necrosis were observed in five patients. CONCLUSIONS: Cryotherapy allows tumor destruction of variable extent in breast carcinomas <2.0cm in diameter. Immunostaining of keratins is useful to identify cytoskeleton remnants of tumor cells in devitalized areas.

Human linear template with mammographic backgrounds estimated with a genetic algorithm.

We estimated human observer linear templates underlying the detection of a realistic, spherical mass signal with mammographic backgrounds. Five trained naïve observers participated in two-alternative forced-choice (2-AFC) detection experiments with the signal superimposed on synthetic, clustered lumpy backgrounds (CLBs) in one condition and on nonstationary real mammographic backgrounds in another. Human observer linear templates were estimated using a genetic algorithm. A variety of common model observer templates were computed, and their shapes and associated performances were compared with those of the human observer. The estimated linear templates are not significantly different for stationary CLBs and real mammographic backgrounds. The estimated performance of the linear template compared with that of the human observers is within 5% in terms of percent correct (Pc) for the 2-AFC task. Channelized Hotelling models can fit human performance, but the templates differ considerably from the human linear template. Due to different local statistics, detection efficiency is significantly higher on nonstationary real backgrounds than on globally stationary synthetic CLBs. This finding emphasizes that nonstationary backgrounds need to be described by their local statistics.

Diagnosis of endometriosis with imaging: a review.

Endometriosis corresponds to ectopic endometrial glands and stroma outside the uterine cavity. Clinical symptoms include dysmenorrhoea, dyspareunia, infertility, painful defecation or cyclic urinary symptoms. Pelvic ultrasound is the primary imaging modality to identify and differentiate locations to the ovary (endometriomas) and the bladder wall. Characteristic sonographic features of endometriomas are diffuse low-level internal echos, multilocularity and hyperchoic foci in the wall. Differential diagnoses include corpus luteum, teratoma, cystadenoma, fibroma, tubo-ovarian abscess and carcinoma. Repeated ultrasound is highly recommended for unilocular cysts with low-level internal echoes to differentiate functional corpus luteum from endometriomas. Posterior locations of endometriosis include utero-sacral ligaments, torus uterinus, vagina and recto-sigmoid. Sonographic and MRI features are discussed for each location. Although ultrasound is able to diagnose most locations, its limited sensitivity for posterior lesions does not allow management decision in all patients. MRI has shown high accuracies for both anterior and posterior endometriosis and enables complete lesion mapping before surgery. Posterior locations demonstrate abnormal T2-hypointense, nodules with occasional T1-hyperintense spots and are easier to identify when peristaltic inhibitors and intravenous contrast media are used. Anterior locations benefit from the possibility of MRI urography sequences within the same examination. Rare locations and possible transformation into malignancy are discussed.

Indeterminate ovarian mass at US: incremental value of second imaging test for characterization--meta-analysis and Bayesian analysis.

PURPOSE: To compare value of current diagnostic strategies in assessment of changes in posttest probability of ovarian cancer when menopausal status and combination and sequence of diagnostic imaging tests are considered. MATERIALS AND METHODS: Prevalence of ovarian cancer according to menopausal status in women with an ovarian mass and performance of combined gray-scale and Doppler ultrasonography (US), computed tomography (CT), and non-enhanced magnetic resonance (MR) imaging and contrast material-enhanced MR imaging after indeterminate results at gray-scale US were derived from meta-analysis by using MEDLINE database and institutional data. Study was approved by the institutional review board of University Hospital Geneva, Geneva, Switzerland; informed consent was waived. Posttest probability values were computed through Bayesian analysis and Monte Carlo simulation after initial gray-scale US and secondary combined gray-scale and Doppler US, CT, or MR imaging, while dependence of test results among imaging modalities was considered. Changes in posttest probability were compared among imaging modalities with summary receiver operating characteristic curves. RESULTS: Prevalence of ovarian cancer was 8.75% in premenopausal women and 32.40% in postmenopausal women with an ovarian mass. After characterization with initial gray-scale US, posttest probability in pre- and postmenopausal women changed, respectively, to 25% and 63% for indeterminate results and to 2% and 7% for benign results. Subsequent use of combined gray-scale and Doppler US, CT, or MR imaging had significant higher positive and lower negative posttest probability than did use of gray-scale US alone. In women with an indeterminate initial US result, posttest probability decreased after secondary testing with benign results for all imaging modalities to 2% in premenopausal women and to 8%-10% in postmenopausal women. After secondary testing for suspicious lesions, posttest probability increased more after non-enhanced (premenopausal women, 70%; postmenopausal women, 92%) or contrast-enhanced MR imaging (premenopausal women, 80%; postmenopausal women, 95%) than it did after combined gray-scale and Doppler US (premenopausal women, 30%; postmenopausal women, 69%) or CT (premenopausal women, 38%; postmenopausal women, 76%) (P < .001). CONCLUSION: In women with an indeterminate ovarian mass at gray-scale US, MR imaging results contributed to change in probability of ovarian cancer in both pre- and postmenopausal women more than did CT or combined gray-scale and Doppler US results.

Ovarian carcinomas in endometriosis: an immunohistochemical and comparative genomic hybridization study.

Malignant transformation of endometriosis, an uncommon phenomenon, can occur in gonadal and extragonadal sites and results in a wide histological range of tumors. Published series reporting malignant transformation of endometriosis have largely been confined to clinical and histopathological discussions with no studies reporting oncoprotein expression and genetic alterations. We report three cases of carcinomas arising in ovarian endometriosis: a serous cystadenocarcinoma, an endometrioid carcinoma with squamous differentiation, and a pure squamous cell carcinoma. Each tumor was analyzed immunohistochemically to compare oncoprotein expression (p53, bcl2, cyclin D1, and c-erb B2) between the tumors and the endometriotic tissue as well as with comparative genomic hybridization (CGH) to compare genetic alterations. All three tumors expressed nuclear p53, in contrast to the endometriotic tissue in which no p53 expression was found. Both endometrial and tumor tissue expressed bcl-2. No expression of cyclin D1 or c-erb B2 was detected in endometriotic or tumoral tissues. The CGH analysis revealed one or two chromosomal aberrations in each of the three tumors with gains on chromosomes 1q, 8q, and 13q, and losses on chromosome 10p. The endometriotic tissue, as expected, showed a normal genetic profile. These results suggest that p53 protein abnormalities and chromosomal aberrations may be involved in malignant transformation of endometriosis in the ovary. However, our results are limited by the number of cases examined and a definite conclusion on the pathogenesis of this process should be followed by future studies with a larger number of cases.