Prognostic and Predictive Factors for Individualized Therapy

Return to Medscape coverage of: 3rd International BCIRG Conference

Prognostic and Predictive Factors for Individualized Therapy

Disclosures

Kathleen I. Pritchard, MD   

Prognostic Factors

Dr. Peter Ravdin,[1] from the University of Texas Health Science Center in San Antonio, Texas, introduced the audience to a number of issues regarding prognostic factors, asking 6 questions: When are prognostic factors needed? What results are obtained? Which prognostic factors do we really need? How useful are these prognostic factors? How do we integrate a variety of prognostic factors?What changes are in store for the future?

According to Dr. Ravdin, prognostic factors are useful only when they contribute information that can affect the clinical decision-making process. To derive good prognostic factors, large data sets of untreated patients with more than 5 years of follow up are needed, and multivariate analysis must be used to examine such data. Prognostic factors that have not been obtained through this process should not be considered pure prognostic factors. A number of interesting data sets (including those from the National Surgical Adjuvant Breast and Bowel Project [NSABP] studies, in which systemic therapy was not given in all arms; the British Columbia data set; the Milan data set; and the Swedish data set) have all shown that grade and tumor size are extremely strong pure prognostic factors. Histologic grade is, in fact, determined by mitotic rate, histologic architecture, and nuclear morphology. Thus, any subset of these factors can also be of prognostic value.

Dr. Ravdin pointed out that HER-2/neu is a relatively strong prognostic factor, but not as strong as some of the more traditional factors such as nodal status and tumor size. The relative risk of recurrence in HER-2/neu-overexpressing node-negative women is 2.2, and the risk of death is 5.5. All studies done so far in node-negative patients show that p53 is a negative prognostic factor, but p53 has never been clearly established as a standard tumor marker because it is measured by a multiplicity of techniques, and standardization has not yet been achieved. Silvestrini and colleagues[2] reported results with p53 showing that it had a relative risk of 1.9 for recurrence and of 2.3 for death.

How should prognostic factors be integrated? Dr. Ravdin has developed a computer program that is available to anyone wishing to use it in their practice. Interested parties should contact Adjuvant_program@aol.com or access http://www.adjuvantsite.com. This program, now used in many centers, allows entry of individual patient factors and offers an overall estimate based on the individual patient's characteristics.

In the future, we may see changes in the availability of prognostic factors as we know them today. For example, with neoadjuvant treatment, strong prognostic factors that are regularly used today, such as number of nodes and grade, may be lost. On the other hand, new techniques such as tumor arrays have been shown in experimental reports to be as predictive as estrogen receptor (ER) and progesterone receptor (PgR) analysis, and indeed to identify specific subgroups of patients who may be more likely to recur or die from breast cancer. These arrays are, however, not yet standardized, and they may require considerable further development before they can be used in routine clinical practice.

Predictive Molecular Markers: Hormone Receptor Status

Dr. Kathy Pritchard,[3] from the Toronto-Sunnybrook Regional Cancer Centre and the University of Toronto, Ontario, Canada, reviewed the role of ER and PgR as prognostic and predictive factors. Prognostic factors are those factors that define the risk of recurrence or death independently of therapy. Predictive factors are those that predict the likelihood of response to a given therapy. ER and PgR are both predictive and prognostic factors.

ER is expressed in 60% to 70% of breast cancers and is a weak but favorable prognostic factor. ER expression provides a 10% to 15% recurrence/survival benefit, as confirmed in at least 4 trials in untreated women. In addition, ER is a predictive factor in women receiving hormonal therapy, with ER-positive women having a 25% to 30% lower chance of developing recurrence. This difference is slightly clearer, larger, and more persistent when ER is measured by immunohistochemistry (IHC) rather than by the older ligand-binding techniques. In addition, ER has also been shown to be of predictive value, since in studies of more than 1000 women in the metastatic disease setting, patients with ER-positive tumors had response rates of up to 70%, while those with ER-negative tumors had response rates of less than 15% to any hormonal therapy.

PgR is an ER-related gene product and indicates whether the ER-regulated pathways are intact. PgR is also both a predictive and prognostic factor. IHC and ligand-binding techniques for PgR are more than 70% concordant. Women with PgR-positive tumors have response rates of more than 70%, while those with ER-negative cancers respond in less than 10% of cases. PgR is a quite weak prognostic factor, with only about a 5% difference in 713 untreated women who have been studied. The disease-free survival in women receiving hormonal therapy is 10% to 20% better if they are PgR positive. Once again, this benefit is more clearly demonstrated with the use of IHC than with the older ligand-binding techniques.

It is important to remember that ER positivity and PgR positivity are associated with a response rate of approximately 77% to hormone therapy, whereas ER- and PgR-negative women have only an 11% response rate. ER-positive/PgR-negative women have a response rate of 27%, whereas ER-negative/PgR-positive patients have a response rate of 46%. Thus, the presence of PgR in the absence of ER, even if it occurs only in about 4% or 5% of patients, strongly suggests that some of these patient are endocrine responsive.

Dr. Pritchard also reviewed data outlining the role of ER as a predictor of response to chemotherapy, a concept that was proposed in the 1970s by Dr. Mark Lippman and colleagues[4] and subsequently rejected due to contradictory data from a series of small studies in patients with metastatic disease. More recently, in the US Intergroup Study of doxorubicin/cyclophosphamide chemotherapy (AC) vs AC followed by paclitaxel (AC/T), an advantage was shown for AC/T, but mainly in ER-negative women. There was little difference in women whose tumors were ER-positive. The Oxford Overview also suggests in some analyses that women with ER-negative tumors many benefit more from chemotherapy.[5,6]

ER and PgR are mechanistic factors rather than only markers. ER is clearly involved in the development and progression of breast cancer, and it is directly linked to the effects of estrogen since it is a nuclear receptor that functions as a transcription factor controlling expression of estrogen-related genes. Ligand binding induces conformational changes that allow interaction of the ER with coregulators, response elements and promotor regions of the target genes (estrogen regulatory elements). All these events contribute to the net estrogenic effect in the target cells.

Polypeptide growth factors and their membrane receptors can also contribute to breast cancer development and progression. Signals delivered through various protein kinase pathways enhance cell survival and proliferation and may interact with the ER-regulated pathways. In fact, many of the new kinases being studied as part of the growth factor cascades can phosphorylate and activate ER, which in turn activates and augments signalling through growth factor pathways. Activation of growth factor pathways may contribute to hormone-resistant states by ligand-independent activation of ER. Thus, targeting of growth factor pathways, in addition to the ER, may provide a double-effect therapy for women with breast cancer.

ER and PgR are still critical factors in the selection of hormonal therapy, and perhaps of other therapies as well. Their measurements must be standardized in terms of tissue preparation, antibody used, scoring, interpretation, and reporting. Unanswered questions include the potential role of ER and PgR in selecting chemotherapy, the role of HER-2/neu in selecting chemotherapy and hormonal therapy, and the role of epidermal growth factor receptor (EGFR) in selecting therapy. In addition, the use of combinations of classic hormonal therapies and biologic agents remains to be evaluated. Optimal guidance for these combined therapies will require carefully standardized laboratory measurements.

HER-2 Status and Response to Therapy

Dr. Mark Pegram,[7] from the UCLA School of Medicine, Los Angeles, California, also evaluated predictive and prognostic factors, particularly HER-2. The growth factor receptor HER-2 is the target of trastuzumab and, thus, overexpression of HER-2 in a cell is predictive of response to the monoclonal antibody. HER-2 is also a prognostic factor in both node-positive and node-negative disease. It is still unclear whether cyclophosphamide/methotrexate/5-fluorouracil (CMF) chemotherapy may not be as active in women who overexpress HER-2/neu. The initial study generated this hypothesis, but did not conclusively substantiate it,[8] and other trials are contradictory in their conclusions. Dr. Pegram reviewed data by Muss and colleagues[9] suggesting that higher-dose cyclophosphamide/doxorubicin/5-fluorouracil (CAF) is better than lower-dose CAF only in women whose tumors overexpress HER-2. However, a subsequent analysis of another subset of patients in the same study by the same authors did not yield consistent results. In addition, transfection of cells with the HER-2/neu oncogene did not change their susceptibility to doxorubicin.[10] Dr. Pegram hypothesized that the actual "culprit" in the association between HER-2 overexpression and response to anthracycline therapy is the enzyme topoisomerase II alpha.

Data are inconsistent also regarding the role of HER-2 and response to taxane therapy. Dr. Pegram's group found that HER-2-overexpressing mice were more sensitive to taxanes, but the opposite has been shown by a group of investigators at MD Anderson Cancer Center. In a clinical cohort of women randomized to receive epirubicin/paclitaxel (ET) vs epirubicin/cyclophosphamide (EC) in metastatic disease, women whose tumors were HER-2 negative had similar responses to the two therapies, while women with tumors overexpressing HER-2 had a much higher response to ET. Survival was also better in ET- vs EC-treated patients only in the HER-2-overexpressing tumors.[11]

In the Breast Cancer International Research Group (BCIRG) 001 Study,[12] treatment with docetaxel/doxorubicin/cyclophosphamide (TAC) polychemotherapy was superior to 5-fluorouracil/doxorubicin/cyclophosphamide (FAC), but more so in Erb B2-overexpressing women in whom the relative risk of recurrence for TAC vs FAC was 0.5 (P = .02) vs 0.74 (P = .06) in women with neu/Erb B2 negative tumors. This difference was, however, not significant.

There seems to be a strong inverse relationship between ER and Erb B2 expression. Most women with ER-positive tumors are non-Erb B2 overexpressors, while many women with ER-negative tumors overexpress neu/Erb B-2. Because ER is a continuous variable and so is HER-2/neu expression, the correlation between these 2 factors alone can explain many of the clinical observations in relation to HER-2/neu and response to hormone therapy. This is not as clear for response to chemotherapy, but it may also be the case.

Predictive Molecular Markers: Topoisomerase II

Dr. Gottfried Konecny,[13] from UCLA, Los Angeles, California, presented data on topoisomerase II alpha and breast cancer. Reviewing the results reported by Paik and colleagues,[14] Di Leo and colleagues,[15] and most recently by Pritchard and colleagues[16] showing superiority of CEF vs CMF in HER-2 overexpressors (P = .06) vs nonoverexpressors (P = .61), he concluded that although these data were quite consistent, coamplification of HER-2 with the topoisomerase II alpha gene may be the factor underlying all these observations.

Topoisomerase II alpha is a key gene in cell division, as it plays an essential role in uncoiling DNA strands and regulating free ends at double strand breaks. Its pattern of expression in HER-2 overexpressors is complex, as both topoisomerase II amplification and deletion may occur. In 100 HER-2-overexpressing patients, 25 were shown to have amplification, 45 normal expression, and 30 deletion of the topoisomerase II gene. However, no topoisomerase II alterations were found in HER-2/neu nonoverexpressors. Thus, these genes seem to be very closely correlated.

Dr. Jarvinen's group in Finland is using a probe for the topoisomerase II gene that is more centromeric than the one used by Konecny and colleagues. Nonetheless, studies by both groups revealed that the amplification of the topoisomerase II gene is strongly correlated with the expression of topoisomerase II alpha protein in cell lines and in patient samples. Lindsay Harris and others[17] have shown that activation of the EGFR and HER-2/neu pathways may induce expression of topoisomerase II, but these results have not been confirmed in other studies.

Isola and colleagues[18] reported that 32% of HER-2 overexpressing tumors also express topoisomerase II. In their work with anthracyclines, tumors with deletion of topoisomerase II had a 17% response rate vs 74% in topoisomerase II overexpressors and 35% in the presence of normal expression. Thus, it may well be that HER-2/neu is not the real "culprit" in anthracycline sensitivity; such sensitivity may be better predicted by accurate measurements of topoisomerase II expression. Dr. Di Leo is now organizing a collaborative group meta-analysis with Dr. Martine Piccart and others that will study a number of tumors with these characteristics to define whether high levels of topoisomerase II alpha do indeed confer anthracyline sensitivity. Since only 25% of breast tumors overexpress HER-2/neu and only a portion of those have a topoisomerase II gene amplification, the added benefit of anthracycline-containing chemotherapies may be restricted to a fairly small subset of breast cancer patients, if these findings are confirmed.

References

Ravdin P. Prognostic factors. Program and abstracts of the 3rd International BCIRG Conference; June 20-22, 2002; Los Angeles, California.Silvestrini R, Daidone MG, Benini E, et al. Validation of p53 accumulation as a predictor of distant metastasis at 10 years of follow-up in 1400 node-negative breast cancers. Clin Cancer Res. 1996;2:2007-2013.Pritchard K. Predictive molecular markers: hormone receptor status. Program and abstracts of the 3rd International BCIRG Conference; June 20-22, 2002; Los Angeles, California.Lippman ME, Allegra J, Thompson EB, et al. The relation between estrogen receptors and response rate to cytotoxic chemotherapy in metastatic breast cancer. N Engl J Med. 1978;298:1223-1228.Cole BF, Gelber S, Coates AS, Goldhirsch A. Polychemotherapy for early breast cancer: an overview of the randomized clinical trials with quality-adjusted survival analysis. Lancet. 2001;358:277-286.Coates AS, Gelber RD, Goldhirsch A. Subsets within the chemotherapy overview. International Breast Cancer Study Group. Lancet. 1998;352:1783-1784.Pegram M. Predictive molecular markers: HER-2 status. Program and abstracts of the 3rd International BCIRG Conference; June 20-22, 2002; Los Angeles, California.Menard S, Valagussa P, Pilotti S, et al Response to cyclophosphamide, methotrexate, and fluorouracil in lymph node-positive breast cancer according to HER2 overexpression and other tumor biologic variables. J Clin Oncol. 2001;19:329-335.Muss HB, Thor AD, Berry DA, et al. c-erbB-2 expression and response to adjuvant therapy in women with node- positive early breast cancer. N Engl J Med. 1994;330:1260-1266.Pegram MD, Finn RS, Arzoo K, Beryt M, Pietras RJ, Slamon DJ. The effect of HER-2/neu overexpression on chemotherapeutic drug sensitivity in human breast and ovarian cancer cells. Oncogene. 1997;15:537-547.Konecny G, Thomssen C, Pegram MD, et al. HER-2/neu gene amplification and response to paclitaxel in patients with metastatic breast cancer. Proc Am Soc Clin Oncol. 2001;20:23a. Nabholtz JM, Pienkowski T, Mackey JR, et al. Phase III trial comparing TAC (docetaxel, doxorubicin, cyclophosphamide) with FAC (5-fluorouracil, doxorubicin, cyclophosphamide) in the adjuvant treatment of node positive breast cancer (BC) patients: interim analysis of the BCIRG 001 Study. Proc Am Soc Clin Oncol. 2002;21:36a. Konecny G. Predictive molecular markers: topoisomerase II expression. Program and abstracts of the 3rd International BCIRG Conference; June 20-22, 2002; Los Angeles, California.Paik S, Bryant J, Tan-Chiu E, et al. HER-2 and choice of adjuvant chemotherapy for invasive breast cancer: National Surgical Adjuvant Breast and Bowel Project Protocol B-15. J Natl Cancer Inst. 2000;92:1991-1998.Di Leo A, Larsimont D, Gangberg D. Her-2 and topoisomerase II alpha as predictive markers in a population of node-positive breast cancer patients randomly treated with adjuvant CMF or epirubicin plus cyclophosphamide. Ann Oncol. 2002; In press.Pritchard KI, O'Malley FA, Andrulis I, et al. Prognostic and predictive value of HER-2/neu in a randomized trial comparing CMF to CEF in premenopausal women with axillary lymph node positive breast cancer (NCIC CTG MA.5). Proc Am Soc Clin Oncol. 2002;21:42a. Harris LN, Yang L, Liotcheva V, et al. Induction of topoisomerase II activity after ErbB2 activation is associated with a differential response to breast cancer chemotherapy. Clin Cancer Res. 2001;7:1497-1504.Isola J, Tanner M, Holli K, Joensu H. Amplification of topoisomerase II alpha is a strong predictor of response to epirubicin-based chemotherapy in HER-2/neu positive metastatic breast cancer. Breast Cancer Res Treat. 2000;64:32.

 
 
This is a part of article Prognostic and Predictive Factors for Individualized Therapy Taken from "Allegra Buy Fexofenadine" Information Blog