| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Imaging, Diagnosis, Prognosis |
Authors' Affiliations: 1 Molecular Oncology Laboratories, Weatherall Institute of Molecular Medicine, University of Oxford, John Radcliffe Hospital, Oxford, United Kingdom; 2 Unità di Patologia Mammaria-Breast Cancer Unit, 3 Centro di Medicina Molecolare ad Indirizzo Oncologico, 4 Anatomia Patologica Azienda Istituti Ospitalieri di Cremona, Cremona, Italy; 5 Department of Pathology, Peter MacCallum Cancer Centre, East Melbourne, Australia; and 6 Oncologia Medica, Dipartimento di Scienze Cliniche e Biologiche, Università di Torino Azienda Ospedaliera San Luigi di Orbassano, Orbassano, Italy
Requests for reprints: Alfredo Berruti, Oncologia Medica, Dipartimento di Scienze Cliniche e Biologiche, Università di Torino, Azienda Ospedaliera San Luigi di Orbassano, Regione Gonzole 10, Orbassano 10043, Italy. Phone: 39-0119026512; Fax: 39-0119026882; E-mail: alfredo.berruti{at}gmail.com.
 |
Abstract |
|---|
 Top Abstract Patients and MethodsResultsDiscussionReferences |
|---|
Experimental Design: PI3K, phospho-AKT (pAKT) and phospho-mTOR were assessed by immunohistochemistry on tumor specimens collected at baseline and after 6 months of treatment in 113 elderly breast cancer patients consecutively enrolled in a randomized phase II trial of primary letrozole therapy and letrozole associated with metronomic cyclophosphamide.
Results: Basal expression of the pathway was not significantly correlated with response or patient outcome. Both letrozole alone and letrozole with cyclophosphamide resulted in a significant reduction of PI3K expression (P = 0.02 and P < 0.005, respectively) and phospho-mTOR expression (P = 0.0001 and P = 0.0001, respectively). pAKT showed no change in the letrozole arm, whereas it was significantly decreased in the letrozole plus cyclophosphamide arm (P < 0.005). pAKT expression reduction was associated with a greater response rate (P = 0.05) and greater reduction in Ki67 expression (P = 0.05). Phospho-mTOR expression reduction was associated with a significantly longer disease-free survival in a multivariate analysis (P = 0.02).
Conclusions: Letrozole inhibits key molecules in the PI3K pathway that are important targets of new drugs being developed to overcome resistance. Changes in these molecules may have prognostic significance. These results should be taken into account when planning prospective trials testing up-front aromatase inhibitor with drugs targeting the PI3K/AKT/mTOR signaling pathway.
However, not all ER-positive breast cancers respond to endocrine manipulation and many initially responding tumors develop resistance. Research into the mechanisms of this endocrine resistance has revealed that cross-talk between estrogen/ER and growth factor signaling through nongenomic interactions enable breast cancer cells to bypass endocrine responsiveness (6).
The growth factor family of epidermal growth factor (EGFR and HER2) are recognized to be implicated in such endocrine resistance (7, 8) through activation of mitogen-activated protein kinase/extracellular signal-related kinase (6, 9, 10) and/or the phosphatidylinositol 3'-kinase (PI3K)/AKT/molecular target of rapamycin (mTOR) pathway. PI3K is a lipid kinase which activates AKT through PDK1. AKT (or protein kinase B) is a serine/threonine kinase that promotes cell survival through the regulation of various downstream effectors, including antiapoptotic pathways. mTOR is a downstream effector of the PI3K/AKT signaling pathway that activates p70S6 kinase and 4E-binding protein-1, which mediate the response of cancer cells to external stimuli and promote cell proliferation, regulating transition through the G1-S phase of the cell cycle (6, 8). In vitro studies have shown that after long-term estrogen deprivation, i.e., during long-term aromatase inhibitor administration, breast tumor cells exhibit an activation of the PI3K/AKT/mTOR pathway as an adaptive phenomenon of breast cancer cells to the low estrogen environment (11).
Two studies in human breast cancer patients have shown that AKT activation is a negative predictor of response to hormone therapy in metastatic breast cancer patients (12) and short survival in nonmetastatic breast cancer patients treated with adjuvant tamoxifen (13). The PI3K/AKT/mTOR pathway, which is often up-regulated in breast cancer (14), therefore represents an attractive target for pharmacologic intervention with signal transduction inhibitors. Approaches to targeting these cell survival pathways have included either specific PI3K inhibitors, such as LY294002, or rapamycin analogues targeting mTOR, such as temsirolimus (CCI-779) or everolimus (RAD-001), both of which are currently under investigation (15, 16).
In vitro studies have shown that breast cancer cells might be particularly sensitive to mTOR inhibitors. In MCF-7 (wt) and MCF-7/Aro (transfected with aromatase gene) breast cancer cell lines, the combination of letrozole and everolimus showed a synergistic interaction in producing a maximal growth inhibition (15). These data suggest that a combined approach may prevent or delay the development of endocrine resistance and could significantly improve the therapeutic efficacy of currently available endocrine options. The efficacy of the combination of aromatase inhibitors with everolimus is currently being tested in several ongoing randomized trials in patients with endocrine-responsive breast cancer (17).
However, theoretical reasons suggest a negative interaction between rapamycin analogues and aromatase inhibitors. If the cross-talk between the estrogen receptors and growth factor signal transduction is highly dependent on ligands, then estrogen deprivation may theoretically down-regulate the PI3K/AKT/mTOR pathway. In addition, estrogen deprivation causes the inhibition of insulin-like growth factor-I synthesis (18), another growth factor involved in the stimulation of the PI3K/AKT/mTOR pathway (19).
To our knowledge, no studies have addressed this issue in human breast cancers in vivo. To investigate the effect of aromatase inhibitors on the PI3K/AKT/mTOR pathway, we have used a primary systemic therapy approach that has proved be an invaluable method for exploring any effect of particular antineoplastic drugs on tumor biology.
We have conducted a randomized phase II trial based on letrozole (LET arm) with or without "metronomic" oral cyclophosphamide. In this trial, the combination of letrozole with cyclophosphamide (LET-CYC arm) resulted in greater tumor shrinkage and greater reduction in either proliferative activity and vascular endothelial growth factor expression compared with letrozole alone (20).
In the present study, PI3K, AKT, and mTOR were assessed on tumor specimens collected before and after treatment in patients randomized in this trial. The primary aim was to explore the changes of these molecular targets before and after treatment, secondary aims were to evaluate the relationship between these targets and conventional clinical and biological prognostic variables and to correlate the changes of these targets with clinical response and patient outcome.
|
Patients and Methods |
|---|
|
TopAbstractPatients and MethodsResultsDiscussionReferences |
|---|
On presentation, an incision biopsy was done on each patient and a small tissue sample (0.5-0.8 cm) was removed. Each month, the size of the primary tumor was measured with a caliper by the same clinician. Response was assessed according to WHO criteria by measurement of the changes in the product of the two largest diameters recorded in two successive evaluations. Tumor progression was defined as an increase of at least 25% in tumor size; stable disease as an increase of <25%, or a reduction of <50%; partial response (4) as a tumor shrinkage >50%; and complete response as the complete disappearance of all clinical signs of disease. Pathologic complete response was defined as the absence of neoplastic cells in the breast and in the axillary lymph nodes after histologic examination.
Surgery was planned after full clinical reassessment. Quadrantectomy or modified radical mastectomy was done when indicated in association with full axillary node dissection. DFS was measured from the date of definitive surgery to the date of recurrence; for patients who did not have a relapse, the data were censored at the date of the last follow-up visit. The study was approved by the local ethical committee. Written informed consent was obtained from all patients before randomization.
Histopathologic grade and immunohistochemistry. All assessments of breast cancer specimens were carried out blinded to patient outcome and whether the samples examined were obtained from incisional biopsy or definitive surgery. Tumor grade was evaluated using the modified Bloom and Richardson system (21). The immunohistochemical methodology used for routine markers as ER, PgR, c-erbB2, ki67, p53, and bcl-2 is fully described elsewhere (20). Tissue microarray reporting tumor from both the initial incisional biopsy and the definitive surgery were used for PI3K mouse monoclonal antibody, phospho-AKT (Ser473) rabbit monoclonal antibody (pAKT), phospho-mTOR (Ser2448) rabbit monoclonal antibody, and phospho-ER
(pER; Ser118; 16J4) mouse monoclonal antibody staining according to the manufacturer's instructions (Cell Signal). These tissue microarrays comprised two 1-mm cores of invasive carcinoma and were scored in a blinded manner on a single occasion using a semiquantitative method as previously described (Fig. 1
; ref. 20). Briefly, the intensity of the staining was scored: (0) no staining, (1) weak staining, (2) moderate staining, or (3) strong staining. All sections had a negative control slide (no primary antibody) of an adjacent section to preclude nonspecific staining. Positive controls included breast carcinomas known to exhibit high levels of each marker.
|
2 test, 2 test for trend, and Fisher exact test were used when indicated to perform comparisons of proportions. Kruskal-Wallis ANOVA was done to compare continuous variables. Comparison of discrete data in matched tumor samples was done using Wilcoxon rank-sum test for paired data. Survival curves were computed according to the Kaplan-Meier method and differences compared using the log-rank test. A multivariate Cox proportional-hazards regression analysis was used to assess the predictive role of changes in phospho-mTOR expression and of clinical and pathologic variables on disease recurrence. The likelihood ratio was used to assess the significance of covariates included in the model. All P values reported were two-sided; values < 0.05 were considered statistically significant. The Statistica for Windows software package was used for statistical analyses.
|
Results |
|---|
|
TopAbstractPatients and MethodsResultsDiscussionReferences |
|---|
|
|
Using a 2 test of variables dichotomized as score 0 and 1 versus score 2 and 3, phospho-mTOR expression directly correlated with pAKT expression (P = 0.012) and PI3K expression directly correlated with pAKT expression (P = 0.01). However, no relationship was observed between PI3K and phospho-mTOR (data not shown). At residual tumor histology, phospho-mTOR expression failed to correlate with either PI3K and pAKT, whereas the relationship between phospho-mTOR and pAKT was maintained (P < 0.004).
Changes in PI3K, pAKT, and phospho-mTOR before and after treatment. A total of 97 patients (45 in the LET arm and 52 in the LET-CYC arm) received definitive surgery (20). Eighty patients (37 in the LET arm and 43 in the LET-CYC arm) had PI3K assessed in matched tumor samples, 85 (39 in the LET arm and 46 in the LET-CYC arm) had pAKT, and 93 (42 in the LET arm and 51 in the LET-CYC arm) had phospho-mTOR assessed. As outlined in Fig. 2 , for PI3K, pAKT, and phospho-mTOR, there were significant reductions in expression in all patients (P = 0.002, P < 0.05 and P = 0.00001, respectively).
|
In detail, 48 cases showed a PI3K reduction and 32 showed no change or increase in PI3K expression. Forty-one cases showed a pAKT reduction, whereas in 24 there was no change or increase in pAKT expression. Phospho-mTOR expression showed a reduction in 62 cases, and no change or an increase in 31.
Dividing patients according to treatment randomization, significant PI3K and phospho-mTOR reduction was observed across both LET (P = 0.02 and P = 0.001, respectively) and LET-CYC (P < 0.005 and P = 0.0001, respectively) arms of the trial. Significant pAKT reduction was observed in LET-CYC cases (P < 0.005) but not in LET cases (P = 0.84; data not shown).
Marker changes in relation with disease response and changes in proliferative activity. Disease response was obtained in 41 out of 57 patients randomized in the LET arm (72.0%) and in 50 out of 57 patients randomized in the LET-CYC arm (88.0%; ref. 20). The relationship between marker changes and disease response is shown in Fig. 3A . For all patients, a reduction in pAKT expression was associated with greater disease response (P = 0.05). For PI3K and for phospho-mTOR, there were no significant relationships to response (P = 0.089 and P = 0.078, respectively). Stratifying patients according to treatment arm showed similar results in the combined series (data not shown).
|
Marker changes and patient outcome. Among the 97 patients receiving definitive surgery, 20 (20.6%) developed disease recurrence after a median follow-up of 46 months. No difference in terms of DFS was observed comparing patients randomized in the two treatment arms (P = 0.52). Basal expression of PI3K (P = 0.57), pAKT (P = 0.22), and phospho-mTOR (P = 0.56) failed to show a predictive role for DFS.
Figure 4 shows the relationship between changes in marker values and DFS. A reduction in phospho-mTOR expression was associated with longer DFS compared with no change or an increase in phospho-mTOR expression. Conversely, changes in PI3K and pAKT failed to be significantly (Fig. 4C) related with DFS. The independent relationship (Fig. 4B) between changes in mTOR expression (Fig. 4A) and DFS was assessed in multivariate analysis after adjustment for tumor stage, grade, lymph node status, primary histology (lobular versus ductal), response to treatment (complete + partial versus no response), and Ki67 expression at posttreatment residual histology (continuous variable). Among the independent variables concomitantly evaluated, phospho-mTOR reduction (hazard ratio, 0.29; 95% confidence interval, 0.1-0.84; P = 0.023; ref. 16) and posttreatment Ki67 expression (hazard ratio, 1.05; 95% confidence interval, 1.003-1.009; P = 0.034) were significantly associated with DFS, respectively, the relationship between disease response and DFS just failed to attain the statistical significance (hazard ratio, 0.34; 95% confidence interval, 0.01-1.16; P = 0.11), whereas the remaining variables did not enter the model.
|
|
Discussion |
|---|
|
TopAbstractPatients and MethodsResultsDiscussionReferences |
|---|
PI3K/AKT/mTOR has been considered a central regulatory pathway of protein translation which is involved in the regulation of cell proliferation, growth, differentiation, migration, survival (23), and paracrine effects such as angiogenesis. Thus, in this study, we evaluated the changes in the PI3K/AKT/mTOR pathway after the administration of letrozole using the primary systemic therapy as an in vivo model.
Baseline PI3K, pAKT, and phospho-mTOR expression did not correlate with the majority of conventional clinical and biological prognostic variables suggesting that this metabolic pathway could potentially provide independent information as previously reported for pAKT (24). PI3K showed an inverse relationship with bcl2 expression and it was more frequently expressed in ductal tumors than lobular tumors. This pattern is consistent with the involvement of this molecule in tumor resistance to endocrine therapy because bcl2 and lobular histology are associated with a greater endocrine sensitivity (25–27). Furthermore, phospho-mTOR expression showed a negative relationship with tumor stage, although this was of borderline significance. This observation is compatible with the induction of phospho-mTOR expression by nutrients (28) and the reduction in nutrient availability with increasing tumor size.
The results also clearly showed that estrogen deprivation down-regulated the PI3K/AKT/mTOR pathway. PI3K and phospho-mTOR, highly expressed at baseline, showed a significant reduction after treatment in all patients and across the two treatment arms. The reduction in PI3K expression after treatment was confined to highly positive cases, whereas the proportion of PI3K-negative cases did not alter. A significant proportion of tumors demonstrating phospho-mTOR positivity before treatment also became negative afterwards.
It is well recognized that endocrine therapy may abrogate cell survival pathways by modifying the insulin-like growth factor receptor and PI3K/AKT pathways. This observation suggests that the efficacy of an aromatase inhibitor could be attributable at least in part to the blockage of metabolic pathways involved in tumor cell survival.
pAKT expression was modestly influenced by treatment with the reduction largely confined to LET-CYC–treated patients, suggesting synergistic activity of the metronomic chemotherapy with endocrine therapy. We have previously shown that the addition of metronomic cyclophosphamide to letrozole resulted in a greater suppression of vascular endothelial growth factor expression (20). Thus, because vascular endothelial growth factor has a growth-stimulatory effect on pAKT, the reduction of pAKT expression may be mediated indirectly through vascular endothelial growth factor down-regulation (29). These findings suggest that a selective and targeted approach combining aromatase inhibitors and signal transduction inhibitors against the above targets should be undertaken.
However, a key observation relates to selective regulation of the pathways by aromatase inhibitors. A significant number of patients have a low expression of the pathways that does not change and would expect little benefit, whereas another group of patients have high levels already suppressed, leaving a group of 
50% of cases for whom it would be worth considering such inhibitors. No studies have thus far considered the differences in pharmacodynamic response as essential for patient categorization and to select for therapy.
In the present study, treatment activity was assessed by either disease response or reduction in Ki67 expression. The validity of clinical response after primary endocrine therapy as a predictor of DFS has been recently questioned (30), whereas changes in Ki67 showed an indirect association with DFS (31, 32). Ki67 expression following primary endocrine therapy residual disease, has recently been shown to provide greater prognostic information compared with Ki67 assessed at baseline (33). In our series, changes in pAKT were significantly associated with a greater response rate and greater reduction in Ki67 expression (P < 0.05). Because in this trial, the combination of letrozole plus cyclophosphamide was significantly more active (both in terms of tumor response and reduction in Ki67 expression) than letrozole alone (20), and pAKT expression reduction was confined in the LET-CYC arm, these data suggest pAKT reduction as a potential mechanism of the greater antitumor activity of the combined arm. Changes in PI3K and phospho-mTOR were not significantly associated with either tumor shrinkage or reduction in proliferation activity. It is noteworthy that phospho-mTOR reduction, despite not being associated with treatment activity, was significantly associated with a longer DFS than no change or an increase in phospho-mTOR. Phospho-mTOR reduction maintained a significant positive prognostic role even after adjustment for commonly recognized prognostic factors, disease response, and posttreatment Ki67 expression, suggesting that the one effect of aromatase inhibitors may be to modify this pathway and the regulation is via a separate mechanism from their antiproliferative effects. The preclinical observations showed that a combination of letrozole and everolimus produced maximal growth inhibition with clear evidence for additive/synergistic effects (15). However, the reduction in mTOR expression after letrozole in the majority of breast cancer cells in vivo and its relationship with good outcome suggest that although the laboratory study merits further investigation, targeting mTOR with CCI-779 or RAD001 requires careful assessment of these pathways before and during therapy. In support of this notion are the data from a phase III trial in advanced breast cancer patients that failed to show any advantage in terms of disease response and DFS by adding temsirolimus to letrozole with respect to letrozole alone (34). However, subpopulations defined by pathways may benefit.
To conclude, the present study shows that letrozole is able to consistently inhibit the expression of key signal transduction and stress response molecules that are important targets of new target therapies as a means to prevent the development of resistance and improve therapeutic efficacy. These results have several implications for trial design: (a) baseline expression of the pathways varies substantially and should be assessed, and (b) reduction in the molecular target should be taken into account and a run-in period of aromatase inhibitor alone is required to classify patients by a pharmacodynamic end point when planning prospective trials with drugs targeting the PI3K/AKT/mTOR signaling pathway. It would then be possible to focus on the relevant subgroup. Our study was done over a 6-month period and it will be of major interest to see if these changes are also induced over short-term exposures of 2 weeks to classify patients even preoperatively. Finally, the independent role of changes in phospho-mTOR before and after treatment in predicting DFS suggests that changes in this marker could provide additional prognostic information to those provided by commonly recognized prognostic markers, treatment response, and changes in proliferative activity.
|
Footnotes |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Received 5/ 1/07; revised 10/22/07; accepted 12/28/07.
|
References |
|---|
|
TopAbstractPatients and MethodsResultsDiscussionReferences |
|---|
, ERBB2, and MAPK signal transduction pathways operate during the adaptation of MCF-7 cells to long term estrogen deprivation. J Biol Chem 2003;278:30458–68.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cancer Prevention Research |
| Cancer Prevention Journals Portal | Cancer Reviews Online |
| Annual Meeting Education Book | Meeting Abstracts Online |
