Contrast-enhanced spectral mammography improves diagnostic accuracy in the symptomatic setting.

AIM: To assess the diagnostic accuracy of contrast-enhanced spectral mammography (CESM), and gauge its "added value" in the symptomatic setting. MATERIALS AND METHODS: A retrospective multi-reader review of 100 consecutive CESM examinations was performed. Anonymised low-energy (LE) images were reviewed and given a score for malignancy. At least 3 weeks later, the entire examination (LE and recombined images) was reviewed. Histopathology data were obtained for all cases. Differences in performance were assessed using receiver operator characteristic (ROC) analysis. Sensitivity, specificity, and lesion size (versus MRI or histopathology) differences were calculated. RESULTS: Seventy-three percent of cases were malignant at final histology, 27% were benign following standard triple assessment. ROC analysis showed improved overall performance of CESM over LE alone, with area under the curve of 0.93 versus 0.83 (p<0.025). CESM showed increased sensitivity (95% versus 84%, p<0.025) and specificity (81% versus 63%, p<0.025) compared to LE alone, with all five readers showing improved accuracy. Tumour size estimation at CESM was significantly more accurate than LE alone, the latter tending to undersize lesions. In 75% of cases, CESM was deemed a useful or significant aid to diagnosis. CONCLUSION: CESM provides immediately available, clinically useful information in the symptomatic clinic in patients with suspicious palpable abnormalities. Radiologist sensitivity, specificity, and size accuracy for breast cancer detection and staging are all improved using CESM as the primary mammographic investigation.

Does the insertion of a gel-based marker at stereotactic breast biopsy allow subsequent wire localizations to be carried out under ultrasound guidance?

AIM: To investigate whether the insertion of a gel-based marker at the time of stereotactic breast biopsy allows subsequent preoperative localization to be performed under ultrasound guidance. MATERIALS AND METHODS: One hundred consecutive women who underwent either a 10 G stereotactic vacuum-assisted breast biopsy or 14 G stereotactic core biopsy with marker placement, followed by wire localization and surgical excision were identified. All had mammographic abnormalities not initially visible with ultrasound. The method of preoperative localization was recorded and its success judged with reference to the wire position on the post-procedure films relative to the mammographic abnormality and the marker. Histopathology data were reviewed to ensure the lesion had been adequately excised. RESULTS: Eighty-three women (83%) had a successful ultrasound-guided wire localization. Successful ultrasound-guided localization was more likely after stereotactic vacuum biopsy (86%) compared to stereotactic core biopsy (68%), although this did not quite reach statistical significance (p=0.06). CONCLUSION: The routine placement of a gel-based marker after stereotactic breast biopsy facilitates preoperative ultrasound-guided localization.

Can computerised tomography replace bone scintigraphy in detecting bone metastases from breast cancer? A prospective study.

BACKGROUND: The aim of this study was to determine whether bone scans (BS) can be avoided if pelvis was included in CT thorax and abdomen to detect bony metastases from breast cancer. MATERIALS AND METHODS: Results of 77 pairs of CT (thorax, abdomen, and pelvis) and BS in newly diagnosed patients with metastatic breast cancer (MBC) were compared prospectively for 12 months. Both scans were blindly assessed by experienced radiologists and discussed at multidisciplinary team meetings regarding the diagnosis of bone metastases. RESULTS: CT detected metastatic bone lesions in 43 (98%) of 44 patients with bone metastases. The remaining patient had a solitary, asymptomatic bony metastasis in shaft of femur. BS was positive in all patients with bone metastases. There were 11 cases of false positive findings on BS. CONCLUSION: Our findings suggest routine BS of patients presenting with MBC is not required if CT (thorax, abdomen, and pelvis) is performed.

NHSBSP type 1 interval cancers: a scientifically valid grouping?

AIM: To assess whether there are differences in the pathological features or survival between the new National Health Service Breast Screening Programme (NHSBSP) interval cancer classification system category of type 1 interval cancers, and the previously used, separate categories of occult, unclassified, and true interval cancers. MATERIALS AND METHODS: The prognostic pathological features (grade, lymph node stage, size, vascular invasion, oestrogen receptor status, and histological type) and survival of 428 type 1 interval invasive breast cancers were analysed by subgroup (occult, unclassified and true interval). RESULTS: Occult cancers compared with other type 1 interval cancers were of significantly lower grade [38 of 52 (73%) versus 151 of 340 (44%) grade 1 or 2, p=0.0005], more likely to be smaller size [37 of 51 (73%) versus 158 of 341 (46%) <20mm, p=0.0003] and more frequently of lobular type at histology [14 of 42 (32%) versus 50 of 286 (17%), p=0.03]. There was no significant difference in pathological features of unclassified tumours compared with other type 1 tumours. There was no significant survival difference between different type 1 subgroups (p=0.12). CONCLUSION: The NHSBSP type 1 interval cancers are a heterogeneous grouping with markedly differing pathological features. However, no significant survival difference is seen between the different type 1 subgroups.

Interval breast cancers: prognostic features and survival by subtype and time since screening.

OBJECTIVES: To investigate the hypothesis that interval cancers arising soon after the previous screen and true interval cancers are biologically aggressive and have a relatively poor prognosis compared with other interval cancers, and to assess which prognostic features are relevant to interval cancers. METHODS: Analysis of prognostic pathological features (grade, lymph node stage, size, vascular invasion, oestrogen receptor [ER] status and histological type), radiological features (comedo/non-comedo calcification and spiculation) and survival for 538 invasive interval breast cancer cases by type and time since previous screen. RESULTS: Late interval cancers were less likely to be lymph node positive (13 versus 43%, P = 0.003). Type 1 interval cancers were more likely to be histological grade 3 than type 2 (minimal signs) and type 3 (false-negative) intervals (52 versus 35%, P = 0.05). Type 3 interval cancers were more likely to have lobular features than other intervals (47 versus 20%, P < 0.0001). There was no significant survival difference by interval cancer type (P = 0.64) or interval year (P = 0.83). At univariate analysis of all interval cancers, tumour size, grade, nodal stage, ER status, vascular invasion and comedo calcification were associated with survival. On multivariate analysis of prognostic features significant at univariate analysis, nodal stage (P value = 0.009), tumour size (P = 0.001), ER status (P < 0.0001) and vascular invasion (P < 0.0001) maintained independent significance. CONCLUSIONS: Our study shows that true intervals and interval cancers arising quickly after screening do not have a worse prognosis than other interval cancers, and that interval cancers have a unique set of prognostic features.

The pathological and radiological features of screen-detected breast cancers diagnosed following arbitration of discordant double reading opinions.

AIM: To compare the mammographic background pattern, the mammographic and the pathological features of screen-detected cancers diagnosed following arbitration of discordant double reading opinions with screen-detected cancers diagnosed following concordant double reading. METHODS: Between April 2002 and December 2003, 249 patients were diagnosed with screen-detected malignancies following concordant double reading. In the same period 38 patients were diagnosed with screen-detected malignancies after their mammograms had undergone arbitration prior to recall. Mammograms of both groups of patients were reviewed retrospectively and the mammographic features documented. Histological data for both groups were also compared. RESULTS: Cancers detected following arbitration were more likely to manifest as parenchymal distortions (44 versus 8%, p<0.001) and less likely to manifest as spiculate masses (19 versus 42%, p=0.014). Arbitration cancers were less likely to be detected in fatty breasts (4 versus 29%, p=0.01). Arbitration cancers were smaller (p=0.045). Lobular cancers were commoner in the arbitration group, although this was of borderline significance (19 versus 8%, p=0.057) There was no significant difference in patient age, tumour grade or lymph node stage between the two groups. CONCLUSION: Cancers detected following arbitration are smaller and more likely to manifest as a parenchymal distortion compared with cancers detected by both readers. Arbitration cancers have broadly similar prognostic features to cancers detected by concordant double reading. It is estimated that approximately 11% more cancers are detected as a result of double reading with arbitration compared with single reading alone, after taking into consideration second reader bias.

Patterns of metastatic breast carcinoma: influence of tumour histological grade.

AIM: To assess if the pattern of metastatic spread of carcinoma of the breast varies according to tumour histological grade. MATERIALS AND METHODS: The clinical details, histological features of the primary tumour, and imaging findings at presentation of patients with metastatic breast cancer have been recorded prospectively since 1997. The pattern of metastatic spread, age at metastasis, metastasis-free interval (MFI), and length of survival with metastases were analysed by tumour grade. RESULTS: There was a significant association between histological high-grade tumours and high frequency of intra-pulmonary metastases (p=0.013); liver metastases (p=0.039); para-aortic lymphadenopathy (p=0.022) and metastatic presentation under 50 years of age (p=0.003). A significant correlation was also demonstrated between histological low-grade tumours and increased frequency of pleural disease (p=0.020); increased frequency of bone metastases (p=0.004); prolonged MFI (MFI>5 years; p<0.0001); and increased length of survival (p<0.0001). CONCLUSION: There is a correlation between patterns of metastatic spread and tumour histological grade. This partly explains the negative prognostic value of high tumour grade, as metastases from grade 3 tumours more commonly occur at sites associated with a worse prognosis. This finding may also prove useful in interpreting imaging in patients who have a history of breast cancer and undergo subsequent imaging because of new symptoms.

Brain metastases from breast cancer: identification of a high-risk group.

AIMS: Brain metastases from breast cancer are an uncommon initial presentation of metastatic breast cancer, but brain metastases commonly occur later in women's metastatic illness. The aims of this study were to document the type, frequency, and temporal occurrence of brain metastases from breast cancer as well as the survival of women with such metastases, and to attempt to identify a subgroup of women at high risk of brain metastases who may benefit from pre-emptive medical intervention. MATERIALS AND METHODS: The radiological reports of all women presenting with metastases aged under 70 years who had subsequently died were examined. The type, frequency, temporal occurrence and survival with brain metastases were documented. Correlations were sought between the frequency of brain metastases and age at metastatic presentation, tumour grade, histological type and oestrogen receptor (ER) status. RESULTS: Of 219 patients who had died with metastatic disease and who were under 70 years of age at metastatic presentation, 49 (22%) developed brain metastases. The development of brain metastases was related to young age (P = 0.0002), with 43% of women under 40 years developing brain metastases. Brain metastases were more common in women whose tumours were ER negative (38%) compared with women with ER-positive disease (14%) (P = 0.0003). By combining age and ER status, it is possible to identify a group of women (age under 50 years and ER negative) with a 53% risk of developing brain metastases. This group included many women who had chemotherapy for visceral metastases, and 68% had either stable disease or disease response at other sites at the time of brain metastases presentation. CONCLUSION: It is possible to identify a subgroup of women with metastatic breast cancer at high risk of brain metastases who may benefit from pre-emptive medical intervention, such as screening or prophylactic treatment.

Woman feels breast lump--surgeon cannot: the role of ultrasound in arbitration.

The role of ultrasound scanning (USS) in patients complaining of a breast lump where the clinical examination (CE) is normal is not clearly defined. To determine this in greater detail, all patients complaining of a breast lump underwent CE. Where no lump could be found, but was still reported by the patient, an USS was performed. All lesions underwent biopsy and/or aspiration, as well as mammography in suspicious cases or those over 40 years of age. This cohort represented 5% of all referrals in the study period. Four hundred and twenty women were prospectively studied in this way. Median follow up is 3.4 years (range 2.5-4.2 years). Twenty two had solid lumps (of which 3 were cancers) and 48 had cysts. Nineteen patients re-presented with symptoms in the same breast (median time = 12 months (range 4.5-20 months), all of which were imaged on USS: 15 cysts and 4 further cancers (3 in the same quadrant as the original lump, one contralateral) were identified. Women with symptomatic breast lumps and a normal CE can be considered a reliable indication that cancer is very unlikely to be present (negative predictive value = 0.98). Ultrasound may be a suitable complimentary investigation, which will relieve symptoms in those with cysts and can detect small clinically--and sometimes mammographically--occult breast cancers.

Metastatic carcinoma of the breast with tubular features: differences compared with metastatic ductal carcinoma of no specific type.

AIMS: To compare the metastatic pattern at presentation and the prognosis with metastases of 48 patients with carcinomas with tubular features (45 tubular mixed and three pure tubular) and 302 patients with tumours of ductal of no special type (DNST). MATERIALS AND METHODS: We carried out a retrospective study from a prospectively maintained database of all patients who developed metastatic disease from carcinoma of the breast in Nottingham, U.K., since 1997. We recorded site of first presentation with metastatic disease, radiological features, histological features and characteristics of the primary tumour. RESULTS: The group of patients with tubular features were older at metastatic presentation (63.9 years vs 59.6 years; P=0.012), had a longer disease-free interval (87 months vs 34 months: P<0.001) and a longer survival with metastases (P<0.002). This group were less likely to have liver metastases (23% vs 41%; P=0.028), in particular multiple liver metastases (50% vs 71%; P=0.015) than the patients with DNST. Other factors known to be associated with prolonged survival, such as low histological grade of the primary invasive tumour and positive oestrogen receptor (ER) status, were more common in the group of patients with tumours with tubular features (Grade 1: 33% vs 3%; Grade 2: 42% vs 25%; Grade 3: 25% vs 72%; P<0.001), (ER positivity 76% vs 52%; P=0.009). When patients with grade 2 tumours were compared, the age at metastatic presentation, disease-free interval and the presence of multiple liver metastases were still significantly different between the two groups. CONCLUSION: Patients with metastatic breast carcinoma with tubular features have a longer survival with metastases than patients with metastatic DNST carcinoma. This improved survival can be explained by better well-recognised prognostic features, such as metastatic site pattern, histological grade, ER status and disease-free interval.

Diagnosis of axillary nodal metastases by ultrasound-guided core biopsy in primary operable breast cancer.

The purpose of this study was to examine the use of ultrasound (US)-guided core biopsy of axillary nodes in patients with operable breast cancer. The ipsilateral axillae of 187 patients with suspected primary operable breast cancer were scanned. Nodes were classified based on their shape and cortical morphology. Abnormal nodes underwent US-guided core biopsy/fine needle aspiration (FNA), and the results correlated with subsequent axillary surgery. The nodes were identified on US in 103 of 166 axillae of patients with confirmed invasive carcinoma. In total, 54 (52%) met the criteria for biopsy: 48 core biopsies (26 malignant, 20 benign node, two normal) and six FNA were performed. On subsequent definitive histological examination, 64 of 166 (39%) had axillary metastases. Of the 64 patients with involved nodes at surgery, preoperative US identified nodes in 46 patients (72%), of which 35 (55%) met the criteria for biopsy and 27 (42%) of these were diagnosed preoperatively by US-guided biopsy. In conclusion, US can identify abnormal nodes in patients presenting with primary operable breast cancer. In all, 65% of these nodes are malignant and this can often be confirmed with US-guided core biopsy.

Detecting which invasive cancers at mammographic screening saves lives?

OBJECTIVES: Assessment of the features of primary operable breast carcinomas arising in women known to subsequently develop metastatic disease, to show for which invasive breast cancers earlier detection by mammographic screening is potentially beneficial. These data were applied to a separate series of screen detected cancers. METHODS: Features associated with the development of metastatic disease after a previous operable breast cancer were ascertained from examination of histological sections of the initial primary carcinoma and in particular the incidence of nodal positivity and definite vascular invasion. Trends in the frequency of nodal involvement and vascular invasion according to histological grade, invasive size, and tumour type were then examined in a further group of 573 screen detected invasive cancers to predict the likelihood of development of systemic disease in these women. RESULTS: Of 173 women who developed metastatic disease after a previous operable breast cancer, 79 (72%) had nodal metastases and 62 (59%) had definite vascular invasion. A high proportion (84%) had either lymph node metastases or vascular invasion or both. The absence of vascular invasion and nodal involvement in invasive breast cancer indicated a low risk of subsequent development of metastatic disease. In the screen detected group, grade 1 invasive cancers <20 mm in size and grade 2 and 3 cancers <10 mm in size had low rates of nodal involvement and vascular invasion. There was a gradual trend to small size, lymph node negativity, and less vascular invasion when comparing screen detected ductal carcinoma of no special type, tubular mixed carcinoma, and tubular carcinoma. Cancers with a lobular component tended to be larger and more often lymph node positive than ductal and no specific type carcinomas. CONCLUSIONS: These data suggest that identification of grade 1 cancers less than 20 mm in size and grade 2 and 3 cancers less than 10 mm in size at screening is likely to be beneficial, with a lower likelihood of developing metastatic disease from these lesions. The detection of tubular mixed carcinomas and ductal carcinomas of no specific type also appears beneficial. However the detection of tumours with lobular features at breast screening does not seem to significantly benefit the patient.

Growth pattern of ductal carcinoma in situ (DCIS): a retrospective analysis based on mammographic findings.

The aim of this study was to obtain information concerning the direction and rates of growth of ductal carcinoma in situ (DCIS). The previous mammograms of 124 women diagnosed with DCIS were examined. If in retrospect calcifications were present on the previous examination, the exact size and position were recorded on both diagnostic and previous imaging. The rates of change and direction of change in extent of calcifications were calculated. 39 women with a diagnosis of DCIS in retrospect had calcifications visible on both their current and prior examinations; these formed the study group. For individual clusters of calcification, change occurred along an axis to the nipple at a mean of 5.5 mm y(-1)and along an axis at 90 degrees to the nipple at 2.6 mm y(-1). Increase in calcifications along the axis to the nipple occurred at 2.6 mm y(-1)toward and 2.8 mm y(-1)away from the nipple. Increase in the axis to the nipple occurred at 1.8 mm y(-1)for low grade, 4.2 mm y(-1)for intermediate grade and 7.1 mm y(-1)for high grade. DCIS growth along an axis to the nipple occurs at over twice the rate of growth in the other direction(s) and growth toward and away from the nipple occurred equally. Growth rates increased with increasing nuclear grade of DCIS. These results validate nuclear grading of DCIS. Additionally, the results suggest that increased importance should be placed on identifying the 'nipple' and 'anti-nipple' margins of DCIS represented by calcifications for both surgical excision and pathological scrutiny.

False-negative breast screening assessment: what lessons can we learn?

AIM: To review women who have had breast cancer diagnosed following previous assessment of a screen-detected mammographic abnormality in order to ascertain the frequency and characteristics of false-negative assessment. MATERIALS AND METHODS: The assessment process was reviewed in the study population of 28 women. This included the nature of the lesion recalled for assessment, additional mammography, clinical and ultrasound findings, and the results of fine needle aspiration cytology and needle histology. RESULTS: The frequency of false-negative assessment was approximately 0.56%. The median time between false-negative assessment and diagnosis of breast cancer was 33 months. The most common mammographic lesion resulting in false-negative assessment was micro-calcification seen in 12 cases (43%). Only five of these 12 cases had image-guided biopsy, the remainder were thought to be benign on magnification views. Other mammographic abnormalities were nine masses (32%), five architectural distortions (18%) and two asymmetric densities (7%). Of the 16 women with mammographic lesions other than micro-calcifications 10 had a normal ultrasound. CONCLUSION: Radiological interpretation of indeterminate micro-calcifications as benign or malignant is unreliable. An isolated cluster of micro-calcification requires image-guided core biopsy with representative micro-calcification obtained on specimen radiography. Further mammography done at assessment, particularly paddle compression views, should be carefully analysed to ensure areas of architectural distortion have truly resolved. If one imaging modality shows a significant abnormality and another does not the cases must be managed on the basis of the abnormal finding. Burrell, H.C.et al. (2001). Clinical Radiology56, 385-388.

Digital imaging improves upright stereotactic core biopsy of mammographic microcalcifications.

AIM: This comparative study was carried out to assess the effect of using digital images compared to conventional film-screen mammography on the accuracy of core biopsy of microcalcifications using upright stereotactic equipment. MATERIALS AND METHODS: The biopsy results from a consecutive series of 104 upright stereotactic 14-gauge core biopsies performed with conventional X-ray (Group A) were compared with 40 biopsies carried out using stereotaxis with digital imaging (Group B). In all cases specimen radiography was performed and analysed for the presence of calcifications. Pathological correlation was then carried out with needle and surgical histology. RESULTS: The use of digital add-on equipment increased the radiographic calcification retrieval rate from 55 to 85% (P < 0.005). The absolute sensitivity of core biopsy in pure ductal carcinoma in situ (DCIS) cases rose from 34 to 69% (P < 0.03), with the complete sensitivity increasing from 52 to 94% (P < 0.005). For DCIS with or without an invasive component the absolute sensitivity rose from 41 to 67% (P = 0.052), while the complete sensitivity was 59% before and 86% after the introduction of digital imaging (P < 0.04). CONCLUSION: Digital equipment improves the performance of upright stereotactic core biopsy of microcalcifications, giving a significantly increased success rate in accurately obtaining calcifications. This leads to an improvement in absolute and complete sensitivity of core biopsy when diagnosing DCIS.

Mammographic features of ductal carcinoma in situ (DCIS) present on previous mammography.

AIM: To review previous mammograms of women found later to have DCIS and identify features which may have been missed or misinterpreted as benign. METHODS: The previous mammograms of 50 women who developed DCIS were analysed. The mammographic features at diagnosis and on the prior mammograms were compared. RESULTS: 11 (22%) of the previous mammograms were in retrospect abnormal; 5 (45%) of these had previously been assessed for the abnormality. All showed microcalcification. The following features were commoner at diagnosis than on previous films; rod shaped calcification (64 vs. 27%, P = 0.03) and a ductal distribution of calcification (76 vs. 45%, P = 0.05). Predominantly punctate calcification (64 vs. 12%, P = 0.001) and less than 10 calcifications in the cluster (54 vs. 24%, P = 0.05) were more common on the previous films. No difference was found in the frequency of granular calcification, branching calcification, irregularity in density, size or shape of calcification between the two groups. CONCLUSION: Features of DCIS missed on previous mammography include small cluster size, less than 10 calcifications in the cluster, the absence of rod shaped calcifications, the absence of a ductal distribution and the presence of predominantly punctate calcification. Features frequently seen both at diagnosis and on previous films which might have allowed earlier diagnosis were granular calcifications which vary in size, density and shape in an irregularly shaped cluster. Focal clustered calcification deserves aggressive investigation.

A comparison of 14 and 12 gauge needles for core biopsy of suspicious mammographic calcification.

This study was carried out to compare the efficacy of 14 vs 12 G needles in stereotactic core biopsy of mammographic calcification. A consecutive series of 100 impalpable mammographic calcifications, without an associated mass and requiring stereotactic core biopsy were randomly allocated to either 14 G or 12 G needle sampling. All biopsies were performed using an upright stereotactic digital unit (Senovision GE) and a Bard automated biopsy gun. Core biopsy results were categorized as either normal, benign, atypical ductal hyperplasia, suspicious of ductal carcinoma in situ (DCIS), DCIS or invasive cancer. The radiographic calcification retrieval rates, complete and absolute sensitivity for malignancy of DCIS and DCIS with an invasive focus were obtained by comparison of core results with surgical histology. Radiographic calcification retrieval was achieved in 86% when using 14 G and 12 G needles. The absolute sensitivity and complete sensitivity for diagnosing DCIS were the same with 12 G and 14 G needles (72% versus 71% and 93% versus 94%, respectively). The use of 12 G needles does not appear to confer benefit over the use of 14 G needles in the diagnosis of mammographic calcification.

Pachymeningitis in Wegener's granulomatosis.

A case of Wegener's granulomatosis is described in which meningeal thickening and enhancement was demonstrated on MR imaging. The diagnosis was suggested by the clinical picture, imaging findings and a positive anti-neutrophil cytoplasmic antibody; and confirmed by biopsy of ethmoidal tissue.

The accuracy of breast ultrasound in the evaluation of clinically benign discrete, symptomatic breast lumps.

INTRODUCTION: Recent reports have suggested that breast ultrasound (US) is of value in distinguishing malignant from benign processes. The aim of this study was to establish the accuracy of US in detecting invasive malignancy in clinically benign, discrete, symptomatic breast lumps. METHODS: The US appearances of 205 clinically benign breast masses were documented prospectively and prior to mammography by one radiologist (AJE). The US appearances were then correlated with the fine needle aspiration (FNA), core biopsy and surgical findings and compared with the mammographic findings. RESULTS: The US findings were normal 72 (35%), simple cyst 63 (31%), solid benign 51 (25%), solid indeterminate 15 (7%) and solid malignant four (2%). Ultrasound characterized 13 (93%) of the 14 patients found to have invasive carcinoma as indeterminate or malignant. No patients with normal or simple cyst US findings had invasive malignancy. Ultrasound had significantly better accuracy (97% vs 87%, P < 0.02) sensitivity (93% vs 57%, P < 0.05) and negative predictive value (99% vs 92%, P < 0.002) than mammography in the detection of invasive carcinoma when indeterminate and malignant imaging findings were taken as positive. CONCLUSION: US is a useful adjunct to FNA/core biopsy in confirming the nature of symptomatic, clinically benign breast masses and is superior to mammography in this clinical setting.