The challenges of integrating molecular imaging into the optimization of cancer therapy.

We review novel, in vivo and tissue-based imaging technologies that monitor and optimize cancer therapeutics. Recent advances in cancer treatment centre around the development of targeted therapies and personalisation of treatment regimes to individual tumour characteristics. However, clinical outcomes have not improved as expected. Further development of the use of molecular imaging to predict or assess treatment response must address spatial heterogeneity of cancer within the body. A combination of different imaging modalities should be used to relate the effect of the drug to dosing regimen or effective drug concentration at the local site of action. Molecular imaging provides a functional and dynamic read-out of cancer therapeutics, from nanometre to whole body scale. At the whole body scale, an increase in the sensitivity and specificity of the imaging probe is required to localise (micro)metastatic foci and/or residual disease that are currently below the limit of detection. The use of image-guided endoscopic biopsy can produce tumour cells or tissues for nanoscopic analysis in a relatively patient-compliant manner, thereby linking clinical imaging to a more precise assessment of molecular mechanisms. This multimodality imaging approach (in combination with genetics/genomic information) could be used to bridge the gap between our knowledge of mechanisms underlying the processes of metastasis, tumour dormancy and routine clinical practice. Treatment regimes could therefore be individually tailored both at diagnosis and throughout treatment, through monitoring of drug pharmacodynamics providing an early read-out of response or resistance.

BRCA1 dysfunction in sporadic basal-like breast cancer.

Basal-like breast cancers form a distinct subtype of breast cancer characterized by the expression of markers expressed in normal basal/myoepithelial cells. Breast cancers arising in carriers of germline BRCA1 mutations are predominately of basal-like type, suggesting that BRCA1 dysfunction may play a role in the pathogenesis of sporadic basal-like cancers. We analysed 37 sporadic breast cancers expressing the basal marker cytokeratin 5/6, and age- and grade-matched controls, for downregulation of BRCA1. Although BRCA1 promoter methylation was no more common in basal-like cancers (basal 14% vs controls 11%, P=0.72), BRCA1 messenger RNA expression was twofold lower in basal-like breast cancers compared to matched controls (P=0.008). ID4, a negative regulator of BRCA1, was expressed at 9.1-fold higher levels in basal-like breast cancer (P<0.0001), suggesting a potential mechanism of BRCA1 downregulation. BRCA1 downregulation correlated with the presence of multiple basal markers, revealing heterogeneity in the basal-like phenotype. Finally, we found that 63% of metaplastic breast cancers, a rare type of basal-like cancers, had BRCA1 methylation, in comparison to 12% of controls (P<0.0001). The high prevalence of BRCA1 dysfunction identified in this study could be exploited in the development of novel approaches to targeted treatment of basal-like breast cancer.

Specific morphological features predictive for the basal phenotype in grade 3 invasive ductal carcinoma of breast.

AIMS: Cytokeratin (CK) 14, a myoepithelial marker, is also expressed in a proportion of breast carcinomas. There is evidence that these tumours show a differing metastatic pattern and clinical outcome from other invasive ductal carcinomas (IDCs) and may need different management. Currently, they are not identified in routine practice and no morphological guidelines exist to aid their identification. The aim of this study was to analyse the histological features of CK14+ IDC. METHODS AND RESULTS: A detailed histological review of 453 grade 3 IDCs revealed 88 (19.4%) that expressed CK14. Assessment was made independently by two pathologists using a standardized 'tick-box' proforma covering grade, architectural and cytological features. The results were analysed using logistic regression to identify features that predicted for basal phenotype. Concordance between the two pathologists was fair to good for most parameters (kappa 0.4-0.6). On multiple logistic regression, the basal phenotype was highly significantly associated with the presence of a central scar (P = 0.005), tumour necrosis (P < 0.0001), presence of spindle cells (P = 0.006) or squamous metaplasia (P < 0.0001), high total mitotic count (> 40 per 10 high-power field) (P = 0.0002) and high nuclear-cytoplasmic ratio (P = 0.0002). CONCLUSIONS: Specific morphological features are strongly associated with basal-like breast carcinoma. These could be used in routine diagnostic practice to identify this important subset of tumours.

Different patterns of inflammation and prognosis in invasive carcinoma of the breast.

AIM: Inflammation in carcinoma of the breast may represent an immune response to the tumour, but there is evidence that this response is impaired. Inflammation may also stimulate tumour growth by releasing proteolytic enzymes and angiogenic factors. Prognostic studies have produced conflicting results, but most investigators have not evaluated the different patterns of inflammation. The aim of this study was to test the hypothesis that moderate or marked diffuse inflammation is associated with a better prognosis. We also tested the 'danger model', which suggests that necrosis is necessary for an effective immune response. METHODS AND RESULTS: On multivariate analysis of women with stage 1 and 2 tumours (n = 679, median follow-up of 9.8 years), survival was independently associated with diffuse inflammation (relative risk 0.43, 95% confidence interval 0.24, 0.77, P =0.005) in addition to histological grade, axillary lymph node status, tumour size and oestrogen receptor status. The presence or absence of tumour necrosis did not have a clear effect on the relationship between survival and diffuse inflammation. CONCLUSIONS: Moderate or marked diffuse inflammation in breast cancer is associated with a better prognosis, suggesting that the immune effects of the inflammation predominate over the protumour effects.

The interaction of oestrogen receptor status and pathological features with adjuvant treatment in relation to survival in patients with operable breast cancer: a retrospective study of 2660 patients.

The oestrogen receptor (ER) status of 2660 patients with primary breast cancer has been related to the effect of different adjuvant systemic therapies on survival. However, as patients in the various treatment groups also had different prognostic features comparison between treatments was difficult. Over 90% of patients receiving tamoxifen (Tam) were postmenopausal compared with <20% of those receiving chemotherapy (CT). The latter had more positive nodes (85% vs 54%) and grade III tumours (54% vs 30%) than the Tam group. The combined CT and Tam group had similar characteristics to the CT alone group. The current reported increase in the proportion of women with ER+ tumours is explained by immunohistochemical analysis of ER and screening programmes. ER status was unrelated to survival in patients with small, low grade, node-negative tumours which was no different from that expected for age-matched women taken from the general population. The value of adjuvant treatment in these patients is therefore questionable. In those given any adjuvant treatment, survival of women with ER+ tumours was prolonged, with the greatest effect being seen in those receiving Tam. Patients with ER- tumours benefited from CT but the addition of Tam to CT improved survival only in those with ER+ tumours. ER status is now established as a major predictive factor for treatment selection in primary disease. Studies of prognostic and predictive markers may be invalidated by use of adjuvant therapy and selection criteria for different treatments. Survival will be influenced by both tumour biology and therapy. This important consideration must be remembered when analysing new markers, particularly in small studies.

Mitotic counts provide additional prognostic information in grade II mammary carcinoma.

The ability to predict how long a patient diagnosed with breast cancer is likely to survive is still imprecise, despite numerous studies which have identified potential prognostic markers. The "established" markers such as nodal status, tumour size, and histological grade have been used for many years and certainly provide some degree of accuracy upon which treatment can be based. However, women with similar prognostic features can vary significantly in their outcome and very few of the newly identified markers provide information that is sufficiently useful to warrant the time and expense spent on their evaluation. In a cohort of 145 women, an assessment has been made of whether knowledge of the proliferative activity of grade II infiltrating ductal breast carcinomas can improve the accuracy of predicting clinical outcome for individual patients. Use of the mitotic count (MC), which was assessed as part of the grading system, enabled patients to be stratified into "good" and "bad" prognostic groups. The measurement of S-phase fraction using flow cytometry gave a similar result, but has the disadvantage that the technique requires specialized equipment. The evaluation of Ki-67 expression using immunohistochemistry was of no additional prognostic value in this defined group. It is proposed that MC, used once to establish grade, could be used again amongst the grade II tumours to improve the accuracy of prognosis and thus influence treatment strategies with minimal additional effort or expense.

The pattern of expression of the microtubule-binding protein RHAMM/IHABP in mammary carcinoma suggests a role in the invasive behaviour of tumour cells.

Intracellular hyaluronic acid binding protein (RHAMM/IHABP), which was recently identified as a novel member of the microtubule-associated protein (MAP) family, has the capacity to interact not only with microtubules but also with microfilaments. The molecule, which is known to be expressed in mammary carcinoma cells, might, through virtue of its intracellular interactions, influence tumour cell morphology and motility. This possibility was examined in a series of 189 mammary carcinomas by immunohistochemistry, using a polyclonal antibody to RHAMM/IHABP. Tumours were selected to include approximately equal numbers of consecutive grade I, II and III ductal carcinomas and invasive lobular carcinomas. Higher grade tumours had significantly lower expression of RHAMM/IHABP in the cytoplasm (p=0.02), but significantly increased expression in trabeculae (p=0.002) and further enhancement at the tumour island edges (p=0.002). Tumours of infiltrating lobular type had stronger expression in the overall cytoplasm (p=0.02) and trabeculae (p=0.08) than carcinomas of ductal type. The presence of strong trabecular expression was associated with a reduced overall survival time (p=0.017). These results suggest that RHAMM/IHABP expression may contribute to the motility and invasiveness of a tumour cell sub-population in breast cancers.

Retention of the expression of E-cadherin and catenins is associated with shorter survival in grade III ductal carcinoma of the breast.

Many studies have investigated the relationship between the E-cadherin/catenin axis and breast cancer biology and yet, unlike the studies in other tumour systems, which have shown a relationship between down-regulation and poor survival, no clear association has emerged in breast. Since accumulating evidence suggests that ductal carcinoma of no special type (NST) represents a diverse group of biologies, this study has focused on grade III ductal carcinoma, in order to reduce the heterogeneity of the study population. A total of 470 breast tumours were studied. Consecutive sections were labelled with antibodies which recognize E-cadherin and the arm proteins with which it interacts: alpha-, beta-, and gamma-catenin. Membrane-bound and cytoplasmic E-cadherin and membrane-bound alpha-catenin expression were associated with a positive oestrogen receptor (ER) status, gamma-catenin with a negative ER status, and, surprisingly, all three with poor survival. Taken together, these findings suggest that a conserved E-cadherin/catenin axis may play a part in determining adverse outcome in grade III breast carcinoma.

Multiple tissue core arrays in histopathology research: a validation study.

The use of multiple tissue arrays allows the examination of large cohorts of tumour tissue with economies of material and technical resources. It also permits the direct comparison of tissues on the same slide. In the present study, a series of 157 breast cancers was labelled with antibodies which recognize oestrogen (ER) and progesterone (PR) receptors and the staining obtained on whole tissue sections was compared with that from a series of multicore arrays. A highly significant association was found between the staining scores (0-7) obtained from the individual tissue sections and from the multicore arrays, although there was some discordance between the receptor status (positive/negative) of the whole section and the tissue core in 5% of cases for ER and in 6.5% of cases for PR. Multiple tissue cores represent an attractive way of dealing with large cohorts of tumours for research studies, because of the significant reduction in reagents and technical time required and the overall speed with which a study can be completed. A proportion of individual tissue cores were not representative of the diagnostic section, which limits the value of multicore arrays as a tool for patient management. However, the technique provides an efficient way of assessing the potential predictive value of novel proteins in different tumour types and in large cohorts.

Molecular markers for predicting prostate cancer stage and survival.

OBJECTIVE: To assess several molecular markers (detected by immunohistochemistry, IHC) to determine whether they can be used to improve the prognostic value of histological grade alone in predicting the behaviour of prostate cancer. PATIENTS AND METHODS: Tumour tissue was retrieved from 156 men in whom tumour grade, stage and survival were known. The outcome measures were: (i) local stage (T-stage, organ-confined vs extraprostatic); (ii) metastatic status (M-stage, bone metastasis vs no bone metastasis); and (iii) survival. The IHC markers used were chosen to provide a broad representation of various aspects of tumour biology, i.e. the androgen receptor (AR) and oestrogen receptor (ER), adhesion molecules (E-cadherin), proliferation markers (MIB-1), tumour-suppressor genes (TP53 and the retinoblastoma gene product, Rb) and other novel cancer-related proteins (cyclin D1 and the breast cancer susceptibility gene product, BRCA2). All factors were assessed using logistic regression and Cox proportional-hazards survival models for predictive value, after adjusting for effects. RESULTS: MIB-1, ER, cyclin D1 and E-cadherin all showed close statistically significant univariate associations with histological grade. Univariate analysis also identified close statistically significant associations between T-stage and both MIB-1 and E-cadherin. Likewise, there were close univariate associations for both M-stage and survival, and MIB-1, cyclin D1 and ER. Logistic regression modelling identified MIB-1, cyclin D1 and ER as statistically significant predictors of M-stage and, once MIB-1 was entered into the model, the effects of grade no longer made a significant contribution. MIB-1 was a significant predictor for T-stage, but the effects of grade remained significant in this model. Cox proportional-hazards modelling identified MIB-1, cyclin D1 and ER as being statistically significant predictors of survival, after adjusting for grade. After adjusting for both grade and MIB-1, the effects of cyclin D1 and ER were no longer statistically significant. Excess MIB-1, cyclin D1 or ER expression tended to be present within the most poorly differentiated and advanced-stage lesions; this provides an inherent instability to the models described. TP53, Rb, AR and BRCA2 were of limited prognostic value. CONCLUSIONS: MIB-1, ER and cyclin D1 provide prognostic information that is clearly independent of grade. However, their true clinical value is probably limited because they are expressed mainly in the most advanced lesions. Nevertheless, MIB-1 expression is of sufficient value to warrant inclusion in future prognostic models. Furthermore, the expression of cyclin D1 and ER may reflect aspects of tumour biology that individually are worthy of further investigation. However, none of the IHC markers used in this study can be recommended for use in routine histological preparations.

Association between loss of heterozygosity of BRCA1 and BRCA2 and morphological attributes of sporadic breast cancer.

Germline mutations in the breast cancer-associated genes BRCA1 and BRCA2 confer a lifetime risk of malignancy. Distinctive morphological features have been attributed to these familial tumours; however, in sporadic breast cancer, the inter-relationship between loss of heterozygosity (LOH) of these loci and tumour morphology remains to be fully elucidated. We studied a series of 120 sporadic breast carcinomas using microsatellite markers to identify LOH of BRCA1, BRCA2, p53 and PTEN. The associations between loss at each of the loci were examined and related to tumour morphology. LOH of the 4 loci did not occur independently; there were highly significant associations between LOH of BRCA1 and both BRCA2 (p < 0.001) and p53 (p < 0.001). LOH at all 4 loci was significantly associated with a high degree of nuclear pleomorphism. Tumours with LOH of BRCA1 also had high mitotic indices, few tubules and a paucity of DCIS, all of which are morphological features similar to those described for familial cases. Following Bonferroni's correction for multiple tests, we found that the tumours with LOH of BRCA1 were still significantly associated with a high mitotic index (p = 0.0006) and a high degree of nuclear pleomorphism (p = 0.001).

Nuclear expression of the c-erbB-4/HER-4 growth factor receptor in invasive breast cancers.

The prevalence and sites of expression of the c-erbB-4 receptor have been determined by immunocytochemical staining in a series of 178 human breast cancers. Most tumors displayed cytoplasmic staining of variable intensity. When compared with adjacent normal tissue, 32 cases (18%) showed lower than normal expression, and 13 (7%) showed greater than normal expression. Nuclear immunoreactivity, confirmed by two different antibodies, was present in 87 cancers (49%) but was found in normal adjacent breast epithelial cells in <5% of cases. There were no significant associations with cytoplasmic or membrane immunoreactivity, but cases showing nuclear expression in >25% of cells were associated with good histological grade, epidermal growth factor receptor expression, c-erbB-3 positivity, cripto, amphiregulin, and transforming growth factor-alpha overexpression.

A role for Id-1 in the aggressive phenotype and steroid hormone response of human breast cancer cells.

The helix-loop-helix protein Id-1 inhibits the activity of basic helix-loop-helix transcription factors, and is an important regulator of cell growth and tissue-specific differentiation. We have shown (P. Y. Desprez et al., Mol. Cell. Biol., 18: 4577-4588, 1998) that ectopic expression of Id-1 inhibits differentiation and stimulates the proliferation and invasiveness of mouse mammary epithelial cells, and that there is a correlation between the levels of Id-1 protein and the aggressiveness of several human breast cancer cell lines. Here, we show that aggressive and metastatic breast cancer cells express high levels of Id-1 mRNA because of a loss of serum-dependent regulation that is mediated by a 2.2-kb region of the human Id-1 promoter. Three lines of evidence suggest that unregulated Id-1 expression may be an important regulator of the aggressive phenotype of a subset of human breast cancer cells: (a) a constitutively expressed Id-1 cDNA, when introduced into a nonaggressive breast cancer cell line (T47D), conferred a more aggressive phenotype, as measured by growth and invasiveness; (b) Id-1 was an important mediator of the effects of sex steroid hormones on T47D cell proliferation. Estrogen stimulated proliferation and induced Id-1 expression, whereas progesterone inhibited proliferation and repressed Id-1 expression. Progesterone repressed Id-1 expression, at least in part by repressing transcription. Most importantly, an antisense oligonucleotide that reduced Id-1 protein levels reduced the ability of estrogen to stimulate cell proliferation, whereas constitutive Id-1 expression rendered cells refractory to growth inhibition by progesterone; and (c) using a limited number of breast cancer biopsies, we showed that Id-1 was more frequently expressed in infiltrating carcinomas compared with ductal carcinomas in situ. Our results suggest that Id-1 can control the malignant progression of breast cancer cells, particularly that mediated by sex steroid hormones. Moreover, Id-1 has the potential to serve as a marker for aggressive breast tumors.

Cell cycle proteins do not predict outcome in grade I infiltrating ductal carcinoma of the breast.

Infiltrating ductal mammary carcinomas are histologically graded according to their extent of differentiation. Well-differentiated, grade I, tumours have low proliferative activity, usually form tubules and exhibit little nuclear pleomorphism. Despite an apparently reassuring morphology, 10-15% of grade I ductal carcinomas metastasize, albeit after a prolonged period. Recent evidence supports the view that evolution to higher grade malignancies occurs rarely and that grade I tumours are biologically distinct from grade III tumours. We have examined a series of 148 grade I ductal carcinomas in order to ascertain whether information about the level of expression of cyclin D1, p27, p53, oestrogen receptor status (ER) or proliferative activity could be used to identify those patients with a poor outcome. The majority of tumours expressed high levels of cyclin D1, p27 and ER, low levels of p53 and had low Ki-67 expression and mitotic counts. Cyclin D1, p27 and ER expression were all significantly correlated with each other but not with p53 (cyclin D1 correlation with ER, p = 0.01; cyclin D1 correlation with p27 and ER correlation with p27 both p < 0.0001). Cyclin D1 and ER were also both correlated with Ki-67 (p = 0.01 and p < 0.0001) but not with mitotic count. Our results suggest that cyclin D1, ER and p27 are all markers of well-differentiated tumours and that their detection is related to proliferative activity in a manner reflecting their functional role within the normal cell cycle. However, none of the proteins or markers of proliferative activity were sensitive enough to predict which patients were likely to have a poor outcome.

Survival of premenopausal breast carcinoma patients in relation to menstrual cycle timing of surgery and estrogen receptor/progesterone receptor status of the primary tumor.

BACKGROUND: Premenopausal breast carcinoma patients who undergo tumor excision during the follicular phase of their menstrual cycle may have a significantly worse prognosis than those whose tumors are excised in other phases of the menstrual cycle. METHODS: Outcome was determined in a series of 112 premenopausal women with operable breast carcinoma in relation to the timing of surgery within the menstrual cycle and the estrogen receptor (ER) and progesterone receptor (PR) status of their primary tumors as determined by immunohistochemistry. RESULTS: Those patients with ER positive tumors who underwent surgery in the early and luteal phase of the cycle had a significantly better survival than women with ER negative tumors (chi-square test = 15.56; P < 0.001). This also was true for PR status (chi-square test = 18.21; P < 0.001). After follicular phase surgery, tumor receptor status had no effect on overall survival. Patients with the best prognosis had ER/PR positive tumors excised on Days 0-2 and 13-32 but even those women with ER or PR negative tumors removed during the luteal phase of their menstrual cycle fared better than patients whose tumors were removed during the follicular phase. CONCLUSIONS: There was a better survival rate for patients with both ER/PR positive and negative tumors treated during the luteal phase of the menstrual cycle. This could be the result of progesterone acting on the surrounding peritumoral normal tissue, thereby exerting a straitjacket effect and improving cohesion of the primary carcinoma. Unopposed estrogen in the follicular phase of the cycle may enable more tumor emboli to escape and successfully establish micrometastases.

The biological and prognostic significance of cell polarity and E-cadherin in grade I infiltrating ductal carcinoma of the breast.

Grading of breast cancer based on the modified Scarff, Bloom, and Richardson system provides invaluable prognostic information. Recent evidence suggests that most tumours do not usually progress between grades and that groups of tumours within each grade are biologically distinct. This study has explored one potential aspect of biological tumour heterogeneity within grade by examining the relationship between cell polarity, the cell adhesion molecule E-cadherin, a major effector of cell polarity, and outcome, in 149 grade I infiltrating ductal breast carcinomas. Polarity was evaluated by studying the degree to which three features of polarized epithelial cells-nuclear ordering, basal positioning of nuclei within cells, and apical snouting/blebbing-were present in these tumours. E-cadherin expression was investigated using the antibody HECD-1. A low degree of tubule formation was correlated with poor nuclear ordering ( p< 0.01). The three histological features-nuclear ordering, basal nuclei, and apical blebbing-were all correlated with each other (all p< 0.0001). Polarity measurements did not correlate with survival. E-cadherin expression did not correlate with polarity and negative tumours were still able to form tubules. Surprisingly, strong E-cadherin immunostaining correlated with poor survival, tumour size, and nodal status. On univariate parametric (Weibull) survival models, high E-cadherin scores and tumour size were both significant predictors of survival in this group.

Effect of c-erbB(2) and estrogen receptor status on survival of women with primary breast cancer treated with adjuvant cyclophosphamide/methotrexate/fluorouracil.

We have investigated the relationship between c-erbB(2) status, estrogen receptor (ER) status and outcome in 274 women with node-positive breast cancer who, following excision and axillary clearance, were randomized to receive either 6 cycles of cyclophosphamide/methotrexate/fluorouracil (CMF) (n = 129) or no such treatment (n = 145). Follow-up data (median 13.3 years) were available on all patients. CMF improved relapse-free and overall survival of all women. The greatest benefit was seen in women with ER-negative tumors; the median overall survival of those given CMF was 11.6 years compared with only 2 years for the control group. For the women with ER-positive tumors the median overall survival of the CMF-treated women was 11.3 years compared with 7.7 years in the control group. When benefit from CMF was examined in relation to c-erbB(2) status, the women with c-erbB(2)-negative tumors who received CMF had a median overall survival of 12.7 years compared with only 7.3 years for the c-erbB(2)-negative women in the control group. The improvement in survival was less marked in the women with c-erbB(2)-positive tumors; median overall survival was 6.1 years for those who received CMF compared with 4.4 years for women in the control group. All women benefited from adjuvant CMF chemotherapy, those with ER-negative tumors benefiting the most. Int. J. Cancer (Pred. Oncol.) 84:354-359, 1999.

Cyclin-dependent kinase inhibitor p27Kip1 expression and interaction with other cell cycle-associated proteins in mammary carcinoma.

p27, cyclin D1, and retinoblastoma (Rb) protein have been demonstrated using immunohistochemistry in 189 cases of primary breast carcinoma with long-term follow-up. There was a statistically significant association between the expression of p27 and both cyclin D1 and the retinoblastoma gene product (pRb), corresponding to their close interactions in regulating the G1/S transition in the cell cycle. Low levels of p27 were seen in high-grade, rapidly proliferating, oestrogen receptor-negative tumours. In univariate analysis, low p27 expression was associated with a reduced relapse-free and overall survival. In multivariate analysis, p27 was not an independent predictor of survival when either histological grade or proliferative activity (S-phase fraction) was included in the model. When the combined expression of p27 and cyclin D1 was related to survival, patients with high levels of p27, regardless of their cyclin D1 status, did well, whilst those with low p27 had a poor outcome. The only exception, in the latter group, was patients with tumours expressing high levels of cyclin D1, who did as well as the high p27 group. We have shown that in clinical material p27 expression is associated with proliferative activity and while univariate analysis shows it to be a significant indicator of prognosis, this significance is lost in multivariate analysis when traditional prognostic factors are included in the model. The interest in p27 expression in mammary carcinoma lies in its behaviour when examined in combination with other G1 cell cycle regulators.