PROGNOSTIC AND PREDICTIVE UPDATE

The science around single and multiplex assays to define prognosis in breast cancer continues to advance rapidly. While new tests are being developed and introduced on a regular basis, it is important to have large and multiple data sets to replicate earlier results in order for clinicians to feel comfortable making treatment decisions that are based on the results of these assays. One of the oldest validated markers is urokinase tissue plasminogen activator/plasminogen activator inhibitor type I (uPA/PAI-1) which is an FDA-approved and guideline-endorsed marker with level 1 evidence. At ASCO, the final results of a large phase 3 trial in node-negative breast cancer utilizing uPA/PAI-1 were presented (1). Patients with low levels of uPA/PAI-1 received no chemotherapy. Those with high levels were randomized to observation or CMF x 6 cycles. One-third were high risk of whom 78% were node negative with 78% also hormone positive, while of the low-risk patients 90% were hormone positive. The ten-year disease-free survival (DFS) for the low-risk group was 87.1% while it was 77% for the high-risk group, p=0.012; for overall survival (OS) the low-risk group was 88.7% at ten years vs. 77.5%.

This marker gave particular prognostic benefit in those with grade 2 tumors. For patients in the per protocol analysis adjuvant CMF reduced disease recurrence by half (HR=0.48, p=0.019). Based of these results, this test represents another useful prognostic assay that could, for example, be utilized for decision making in individual patients with intermediate Oncotype DX® recurrence scores in whom the benefit of CMF chemotherapy is not clear, or in patients with hormone receptor-negative, lymph node-negative tumors who are borderline candidates for chemotherapy.

The 70-gene assay, commercially available as MammaPrint, divides patients into low risk and high risk for distant disease-free survival (DDFS) and is a prognostic factor in T1-2 breast cancer with either lymph nodes negative or positive and with either ER positivity or negativity. Its value in also functioning as a predictive marker for the benefit of adjuvant chemotherapy was the subject of an oral presentation at ASCO, based on a retrospective pooled analysis of previous studies totaling 1600 patients (2). Half of these subjects were node positive and had >T1 tumors. Forty percent received hormone therapy and chemotherapy while 60% received hormone therapy alone for estrogen receptor-positive tumors. The patients were sorted into low and high risk categories. The 5-year DFS for low-risk patients was 95% vs. 82% for high-risk patients (HR=3.88, p=<0.01). Those patients who received chemotherapy in the low-risk group did not seem to benefit although they did numerically better. In the high-risk group, hormonal therapy plus chemotherapy improved 5-year DDFS to 88% from 76%.

This study is weakened by the nonrandomized approach to the treatment decision of the addition of chemotherapy. A validation of the 70-gene assay against a prior study that randomized patients to chemotherapy or none would be useful. Alternatively, the prospective study MINDACT, currently in progress, will attempt to validate the use of this assay for making chemotherapy treatment decisions. MammaPrint requires fresh frozen tissue rather than archived formalin-fixed paraffin-embedded specimens and therefore is limited on a practical basis. However, this test which is also FDA-approved can be considered another useful aid in helping to determine risk when circumstances require more information.

ADJUVANT HORMONAL THERAPY

The meta-analysis of aromatase inhibitors (AIs) as adjuvant therapy presented at San Antonio Breast Cancer Symposium in December 2008 (3) confirmed the overall benefit of AIs over tamoxifen in both the initial treatment of hormone receptor-positive postmenopausal breast cancer and as part of a switching strategy for patients after 2-3 years of tamoxifen. The proportional benefit for DFS is approximately 15% which translates into an absolute benefit of 3-5%. However, there is no overall survival benefit demonstrated even in the meta-analysis and for multiple reasons, many women continue to be prescribed tamoxifen at some point in their course of adjuvant hormonal therapy. Therefore, it is critical to continue to explore determinants of tamoxifen (and AIs for that matter) efficacy in order to improve upon the 50-65% reduction in breast cancer recurrence documented with these drugs.

It has been recognized for several years that tamoxifen is a pro-drug and has to be converted through two enzymatic steps into the molecular entity that exhibits the pharmacologic effect on estrogen receptor inhibition. Tamoxifen is metabolized by cytochrome P450 enzymes, most critically the CYP2D6 enzyme which is primarily responsible for the conversion of tamoxifen to endoxifen, the active species. From a pharmacogenetic standpoint, it is known that approximately 10% of the Caucasian population has genetic polymorphisms in 2D6 rendering those poor metabolizers. As a result, they produced low levels of endoxifen from tamoxifen. Some but not all studies suggest that poor metabolizers of tamoxifen have an increased risk of breast cancer recurrence.

Likewise, there are many drugs which can induce or inhibit the activity of 2D6 and earlier literature has variously suggested that such drugs may also influence endoxifen production and alter breast cancer outcome. Studying this effect requires large populations and prospective studies have not been and are not likely to be performed. Since the commonly prescribed SSRI antidepressants paroxetine and fluoxetine are moderate/severe inhibitors of 2D6, this question has practical relevance for many women who are taking these medications alongside tamoxifen.

So it was with great interest that the adjuvant oral session at ASCO presented two methodologically similar retrospective studies which attempt to answer the question of whether drug interactions affect breast cancer recurrence in patients taking tamoxifen. The first study, conducted using the MEDCO claims database included women taking tamoxifen for at least 2 years and also included those taking no medications that inhibit 2D6, weak inhibitors and moderate/strong inhibitors (4). The groups were matched for age, surgery and adjuvant treatment. Results from this study showed a near doubling of recurrence rate for patients receiving moderate/strong inhibitors of 2D6 compared to no inhibition, 14.0% vs. 7.5% recurrence rates (HR=1.92, p=<0.001), including the SSRIs in the inhibitor category.

The second study, performed in the Netherlands by evaluating three national databases, used similar but not identical methodology to evaluate concurrent usage of tamoxifen plus 2D6 inhibitors and included fewer patients receiving an inhibitor, most taking these before the calendar year 2001 (5). They demonstrated no significant difference in outcome between those patients receiving a moderate/strong 2D6 inhibitor vs. none and could not show a difference even when restricting analysis to the strong inhibitor group only. However, in a secondary endpoint they demonstrated an increased risk for recurrence in patients whose adherence with tamoxifen was <90%.

What is the take home lesson from these data? The jury is still out on whether CYP2D6 plays a clinically significant role in the outcome of patients on tamoxifen whether considered based on endogenous metabolism or exogenous concomitant drug usage including SSRIs which exhibit moderate to strong inhibition of drug metabolism. It seems prudent, given the alternatives available for both endocrine therapy and for anti-depressants, to utilize non-inhibitors of 2D6 whenever possible and to consider testing for intrinsic 2D6 metabolism by genetic analysis in cases where inactivity of tamoxifen will have significant impact. Clearly we need more studies to address this important aspect of hormonal therapy.

METASTATIC BREAST CANCER

This year’s ASCO breast cancer track arguably generated the most excitement around new developments in the treatment of metastatic breast cancer. A good deal of that enthusiasm surrounded the studies utilizing PARP inhibitors to target breast cancer cells. In order to understand how PARP inhibitors work, it is useful to review biological aspects of two subsets of breast cancer: those with triple negative breast cancer (TNBC) and those occurring in patients with a BRCA1-2 mutation.

BRCA MUTATION RELATED BREAST CANCER

Tumors in BRCA mutation carriers lack functioning BRCA protein. Both BRCA1 and 2 are associated with DNA repair. The hypothesis has arisen that aberrant or absent BCRA1 function might confer sensitivity to DNA damaging agents such as platinum salts. Importantly, the absence of BRCA repair mechanisms might increase tumor cell reliance on other important DNA repair systems such as PARP (Poly (ADP-ribose) polymerase) which is a key regulator of DNA repair processes through base excision repair. By inhibiting this enzyme, homologous recombination of DNA may not occur leading to the inability to replicate cells accurately and ultimately tumor cell death.

TRIPLE NEGATIVE BREAST CANCER

TNBC refers to tumors with an immunohistochemically defined phenotype which is negative for ER/PR and HER2 expression. There is a great deal of overlap of TNBC with the genomic classified basal subtype of breast cancer, although one quarter of basal cancers are not TNBC and one-quarter of TNBC do not fall into the basal category (6). TNBC is associated clinically with a higher risk of relapse, visceral involvement and the lack of a specific targeted therapy. TNBC shares several features with BRCA1-related breast cancers: 80% of BRCA1 cancers are negative for ER/PR and HER2, both typically have mutant P53 genes, look basal-like on gene expression, and have high grade tumors histologically. Moreover, PARP concentrations tend to be upregulated in both TNBC and BRCA1 mutated breast cancers. Both TNBC and BRCA cancers tend to be sensitive to DNA damaging agents. These similarities lead to both groups being strong candidates for treatment with PARP inhibitors.

PARP INHIBITORS PLUS DNA-DAMAGING AGENTS IN TRIPLE NEGATIVE BREAST CANCER

The plenary session at ASCO included a randomized phase 2 study of the small molecule IV PARP inhibitor, BSI-201, along with chemotherapy in patients with TNBC (7). This relatively small trial randomized 120 patients to gemcitabine and carboplatin vs. gemcitabine, carboplatin and BSI-201. Most patients had had prior exposures to taxanes and anthracyclines. PARP was found to be upregulated in most TNBCs. The clinical efficacy results were striking: a 48% response rate in the BSI-201 group vs. 16% in the chemotherapy only group, p=0.002, and a clinical benefit rate of 62% vs. 21%, p=0.0002. The progression free survival for the BSI-201 group was 6.9 months vs. 3.3 months for chemotherapy only, HR=0.342, p=<0.0001 and in an exploratory analysis there was even an increase in overall survival to 9.2 months for the BSI group vs. 5.7 months for chemotherapy only, HR=0.348, p=0.0005. Moreover, there was little if any added toxicity to BSI-201 combined with chemotherapy. These extraordinarily promising results will be evaluated in a phase 3 trial with a similar study design soon to begin.

PARP INHIBITORS IN BRCA-ASSOCIATED BREAST CANCER

Also exciting was a phase 2 trial of an oral small molecule PARP inhibitor called olaparib, which has previously shown proof of concept with activity in a recently reported phase 1 trial which included BRCA1-mutated ovarian cancer patients(8). In this sequential cohort study, 27 patients were treated at the previously determined MTD while a second cohort of 27 received a reduced dose at one-quarter of the MTD (9). All patients had received taxanes and anthracyclines and the median number of prior chemotherapy regimens was three. The overall response rate for the high-dose single agent oral therapy was 11/27 or 41% while in the low-dose cohort it was 22%. Both BRCA1 and 2 mutation carriers responded. The PFS for the high dose group was 5.7 months. Olaparib was generally well tolerated with fatigue, nausea, vomiting, headache, constipation and diarrhea being the most common symptoms. Less than 20% of patients had grade 3 fatigue, nausea or vomiting. These results will be confirmed in a larger BRCA mutation phase 2 trial.

Another approach to BCRA mutation carriers was presented by a group from Poland who used single agent cisplatin as neoadjuvant treatment in 25 patients with BRCA1 mutations (10). After four cycles of platinum therapy, all patients underwent a mastectomy and 18/25 (72%) had a pCR in breast and lymph nodes. This small study suggests that further clinical trials of platinum-based chemotherapy are warranted to explore the DNA damage sensitivity of BRCA mutated tumors.

OTHER TOPICS IN METASTATIC BREAST CANCER

Scheduling of drugs in MBC remains an issue of concern, particularly as it relates to doublet therapy and best first agent(s) choice. Seidman, et.al. presented a randomized trial of gemcitabine and docetaxel followed by capecitabine compared to capecitabine and docetaxel followed by single agent gemcitabine (11). The response rates in both first line regimens were similar as were the progression-free survival in both the induction and crossover phases; although, the gem/docetaxel followed by capecitabine appeared numerically somewhat better. At least half the patients in both arms required dose reductions. Given the sensitivity of breast cancer to multiple types of chemotherapy, this study suggest the choice of agent is less important than making sure patients are exposed to multiple active agents over the course of their treatment.

Combination therapy consisting of chemotherapy plus bevacizumab, an anti-angiogenic monoclonal protein targeting VEGF, has been shown to improve PFS in the first line treatment of MBC when combined with either paclitaxel or docetaxel. However, OS has not been increased with bevacizumab. A third phase III trial investigating the benefit of chemotherapy plus bevacizumab was reported by Robert et.al. at ASCO 2009 on behalf of the RIBBON-1 investigators (12). In this trial, investigators could choose from among several different agents to combine with bevacizumab. The PFS and response rate was significantly improved when bevacizumab was added to taxane, anthracycline or capecitabine treatment. While adverse events were higher in the combination therapy arms, no new safety signals were identified. This study confirms the benefit of adding bevacizumab when chemotherapy is initiated for MBC.

Another persistent question in treatment of metastatic breast cancer has been the benefit of maintenance chemotherapy after induction treatment in the first line setting. Put another way, is it safe to allow patients a chemotherapy holiday after induction treatment? Mayordomo and coauthors from Spain randomized patients to receive six cycles of sequential epirubicin and paclitaxel or the same followed by low dose weekly paclitaxel until progression or unacceptable toxicity (13). Randomization occurred prior to induction. The median PFS was twelve months in the maintenance arm vs. eight months in the induction only arm, which was not statistically significant. Importantly, the median overall survival was 24 months in both arms.

A study reviewing phase 3 trials published this decade in major journals found that there had been no overall increase in median overall survival over the study (14). Survival did correlate with gains in PFS but the median post-progression survival has a stronger correlation with the overall survival than the PFS from the induction regimen of interest. These findings suggest that subsequent lines of therapy contribute strongly to overall survival.

Taken together these presentations illustrate the need for careful individual selection of first and subsequent therapies for patients diagnosed with metastatic disease. The goal should be maintenance of both quality and quantity of life with limiting disease-related symptoms and treatment-related side effects.

CONCLUSION

Clinical advances in breast cancer are proceeding through at least three pathways: 1) by fine tuning the extensive treatment options already available, 2) refining prognostic and predictive tools to allow the most effective therapies to be fitted to those who will need them and benefit most, and 3) by validating new targets through translational research. All three approaches were in abundant evidence at ASCO 2009. The progress seen year to year at this meeting translates into a tangible benefit for patients as evidenced by improvement in survival rates over this decade. We look forward to continued evolution of the treatment of breast cancer in years to come.

REFERENCES

1. Thomssen C, Vetter M, Schmidt M, et al. ASCO-recommended prognostic factors uPA/PAI-1 in node-negative (N0) breast cancer patients (pts) compared to clinicopathological risk assessment within the NNBC 3-Europe trial. Program and abstracts of the 2009 Annual Meeting of the American Society of Clinical Oncology, Orlando, Florida (Abstract 544).

2. Bender RA, Knauer M, Rutgers EJ, et al. The 70-gene profile and chemotherapy benefit in 1,600 breast cancer patients. Program and abstracts of the 2009 Annual Meeting of the American Society of Clinical Oncology, Orlando, Florida (Abstract 512).

3. Ingle JN, Dowsett M, Cuzick J, et al. Aromatase inhibitors versus tamoxifen as adjuvant therapy for postmenopausal women with estrogen receptor positive breast cancer: meta-analyses of randomized trials of monotherapy and switching strategies Cancer Research 2009; 69:66s, Abstract 12.

4. Aubert RE, Stanek EJ, Yao J, et al. Risk of breast cancer recurrence in women initiating Tamoxifen with CYP2D6 inhibitors. Program and abstracts of the 2009 Annual Meeting of the American Society of Clinical Oncology, Orlando, Florida (Abstract CRA508).

5. Dezentje V, Van Blijderveen NJ, Gelderblom H. Concomitant CYP2D6 inhibitor use and tamoxifen adherence in early-stage breast cancer: a pharmacoepidermiologic study. Program and abstracts of the 2009 Annual Meeting of the American Society of Clinical Oncology, Orlando, Florida (Abstract CRA509).

6. Parker JS, Mullins M, Cheang MC, et. al. Supervised risk prediction of breast cancer based on intrinsic subtypes. J Clin Onc 2009;27:1160-1167.

7. O’Shaughnessy J, Osborne C, Pippen J, et al. Efficacy of BSI-201, a poly (ADP-ribose) polymerase-1 (PARP1) inhibitor, in combination with gemcitabine/carboplatin (G/C) in patients with metastatic triple-negative breast cancer (TNBC): results of a randomized phase II trial. Program and abstracts of the 2009 Annual Meeting of the American Society of Clinical Oncology, Orlando, Florida (Abstract 3).

8. Fong PC, Boss DS, Yap TA, et al. Inhibition of poly(ADP-Ribose) polymerase in tumors from BRCA mutation carriers. N Engl J Med 2009; 361(2):1-12.

9. Tutt A, Robson M, Garber JE, et al. Phase II trial of the oral PARP inhibitor olaparib in BRCA-deficient advanced breast cancer. Program and abstracts of the 2009 Annual Meeting of the American Society of Clinical Oncology, Orlando, Florida (Abstract CRA501).

10. Gronwald J, Byrski T, Huzarski T, et al. Neoadjuvant therapy with cisplatin in BRCA1-positive breast cancer patients. Program and abstracts of the 2009 Annual Meeting of the American Society of Clinical Oncology, Orlando, Florida (Abstract 502).

11. Seidman AD, Brufsky A, Ansari RH, et al. Phase III trial of gemcitabine plus docetaxel (GD) compared to capecitabine plus docetaxel (CD) with planned crossover to the alternate single agent in metastatic breast cancer (MBC). Program and abstracts of the 2009 Annual Meeting of the American Society of Clinical Oncology, Orlando, Florida (Abstract 1000).

12. Robert NJ, Dieras V, Glaspy J, et al. RIBBON: Randomized, double-blind, placebo-controlled, phase III trial of chemotherapy with or without bevacizumab (B) for first-line treatment of HER2-negative locally recurrent or metastatic breast cancer (MBC). Program and abstracts of the 2009 Annual Meeting of the American Society of Clinical Oncology, Orlando, Florida (Abstract 1005).

13. Mayordomo J, Baena J, Ciera L, et al. Final results of a randomized trial on the role of maintenance chemotherapy with weekly paclitaxel for patients with metastatic breast cancer. Program and abstracts of the 2009 Annual Meeting of the American Society of Clinical Oncology, Orlando, Florida (Abstract 1001).

14. Katz A, Saad ED, Buyse ME, et al. Overall survival (OS) in contemporary randomized clinical trials (RCT) in advanced breast cancer (ABC). Program and abstracts of the 2009 Annual Meeting of the American Society of Clinical Oncology, Orlando, Florida (Abstract 1002).