Important Trials of the Last Decade

Breast cancer is the most common cancer in women worldwide, and is the second leading cause of cancer death in US women. Breast cancer mortality has declined by approximately 40% over the past several decades in the United States due to widespread mammographic screening and use of adjuvant systemic therapies for those with localized disease. Several studies performed in the 1980s and 1990s demonstrated that multidrug combination chemotherapy regimens reduce breast cancer recurrence and mortality, and that proportional reductions in recurrence risks occur irrespective of age, nodal status, tumor size, tumor grade, and hormone receptor expression. These findings led to an expert NIH panel to recommend use of adjuvant chemotherapy not only in patients with node-positive disease at high risk of recurrence, but also with patient at lower risk of recurrence with node-negative, hormone receptor–positive disease and tumors larger than 1 cm, resulting in overtreatment of a subgroup of patients who might otherwise have been cured without adjuvant chemotherapy.

Over the last decade, several trials have evaluated the role of gene expression assays to tailor the use of adjuvant chemotherapy in hormone receptor–positive, HER2-negative localized breast cancer. In addition, clinical trials evaluating the role of various systemic therapeutic approaches have been shown to improve clinical outcomes in locally advanced and/or metastatic breast cancer, including anti-HER2 directed antibodies and tyrosine kinase inhibitors in HER2-positive breast cancer, antibody-drug conjugates for HER2-positive and triple-negative breast cancer, cyclin-dependent kinase (CDK) 4/6 inhibitors for hormone receptor–positive breast cancer, and antibodies directed at immune checkpoints including programmed cell death 1 (PD-1) and programmed death ligand 1 (PD-L1) for triple-negative breast cancer. This chapter summarizes several of these important trials in localized, locally advanced, and metastatic breast cancer.

Trials Evaluating Deescalation of Chemotherapy

Trials Evaluating Use of Gene Expression Assays to Deescalate Chemotherapy

Multiparameter gene expression assays measure RNA expression of multiple genes and combine them into a signature to evaluate prognosis of breast cancer. Prospective-retrospective studies using archived tumor tissue have demonstrated that several of these assays are prognostic for breast cancer recurrence, and, in some cases, predictive of benefit from adjuvant chemotherapy or extended adjuvant endocrine therapy (ET), providing level 1B evidence supporting their use for these indications. Two assays, the 70-gene MammaPrint assay, and the 21-gene Oncotype DX Recurrence Score (RS), have been evaluated in large prospective randomized trials required to provide level IA evidence supporting their use for tailoring of adjuvant chemotherapy in hormone receptor–positive breast cancer with up to three positive axillary nodes (summarized in Table 3.1 ).

Table 3.1

Prospective Studies Evaluating Gene Expression Assays for Tailoring Adjuvant Chemotherapy

Trial	n	Population	Randomization	Primary End Point	Key Findings
MINDACT	6693	HR+, HR−, HER2+, HER2−, N0, N1	Adjuvant therapy based on clinical risk vs. on genomic risk for patients with discordant clinical and genomic risks by MammaPrint assay	DMFS in clinical high-risk/genomic low-risk group	Clinical high- and genomic low-risk disease may be spared chemotherapy Small chemotherapy benefit in women aged <50 years with high clinical risk and low genomic risk Concordant clinical and genomic low risk associated with excellent DMFS
TAILORx	10,273	HR+, HER2−, N0	CET vs. ET for patients with 21-gene RS 11–25	iDFS	ET noninferior to CET for RS 11–25 Small chemotherapy benefit in women ≤50 years with RS 21–25 or high clinical risk and RS 16–20 Low DR with ET alone for RS 0–10 Patients with RS ≥26 have better outcomes with CET than expected for ET alone
RxPONDER	5083	HR+, HER2−, N1, RS 0–25	CET vs. ET	iDFS; interaction of chemotherapy and RS	No chemotherapy benefit in postmenopausal women, regardless of RS 5.2% improvement in iDFS with CET in premenopausal women, regardless of RS

CET , Chemoendocrine therapy; DFS , disease-free survival; DMFS , distant metastasis-free survival; DR , distant recurrence; ET , endocrine therapy; HR+ , hormone receptor–positive; HR− , hormone receptor–negative; HER2+ , HER2-positive; HER2− , HER2-negative; iDFS , invasive disease-free survival; N0 , node negative; N1 , one to three axillary nodes involved; OS , overall survival; RS , recurrence score; T , tumor size.

Mindact

The 70-gene MammaPrint assay evaluates RNA expression of 70 genes and classifies tumors as having a low or high risk of recurrence, with low genomic risk corresponding to a 5-year distant metastasis-free survival of >90%. This assay was prospectively evaluated in the MINDACT trial, which enrolled 6693 women with early breast cancer (HR+ and HR–, and HER2+ and HER2–) and up to three involved axillary nodes. Clinical risk of recurrence was calculated using the Adjuvant!Online algorithm (for HR+ HER2− disease, clinical low risk is defined as node-negative tumors up to 1 cm in size and high-grade, 2 cm in size and intermediate-grade, or 3 cm in size and low grade). Genomic risk of recurrence was assessed by MammaPrint. Patients with concordant high clinical and genomic risk of recurrence (27%) were recommended to receive chemotherapy, whereas patients with concordant low risk (41%) were advised against chemotherapy. Patients with discordant risks (34%) were randomized to treatment based on genomic risk or clinical risk. The primary aim of the trial was to determine if the subgroup of patients with clinically high-risk but genomically low-risk disease who were randomized to treatment by genomic risk (no chemotherapy, n = 644) had a 5-year distant metastasis-free survival (DMFS) with a lower 95% CI boundary of at least 92%. This subgroup of patients had fairly heterogeneous clinicopathologic characteristics: while 97.5% were HR+, 8% were HER2+, and 48% had involvement of axillary nodes. DMFS for this group was 94.7% (95% CI 92.5%–96.2%), meeting the study’s primary end point. Clinical high-risk/genomic low-risk patients receiving chemotherapy had a 1.5% (SE + 2.3%) improvement in DMFS over those who did not; however, MINDACT lacked sufficient power to determine chemotherapy benefit in the low genomic risk group. Updated analyses of MINDACT showed that the 8-year DMFS for clinical high/genomic low patients was 89.4% (95% CI 86.8–91.5) without chemotherapy versus 92.0% (95% CI 89.6–93.8) with chemotherapy, and that chemotherapy benefit was limited to patients aged 50 years or under. MINDACT patients with low clinical risk demonstrated an excellent 8-year DMFS (94.7% [95% CI 93.8%–95.6%]) for concordant clinical and genomic low-risk patients, with a 3.6% DMFS decrement associated with high genomic risk. In addition, a subgroup of 509 patients with clinically low-risk disease, all of whom had tumors that were HR+, node-negative, and under 2 cm in size, and 95% of whom were also genomic low-risk, received no adjuvant systemic therapy at all, and demonstrated 8-year DMFS of 94.8% (95% CI 92.7–96.9). In contrast, concordant clinical high/genomic high-risk patients had substantially poorer outcomes, with 5- and 8-year DMFS of 90.6% (95% CI 89.1–91.9) and 85.9% (95% CI 84.2–87.5).

TAILORx

The 21-gene Oncotype DX assay evaluates expression of 16 cancer-related genes and five reference genes by RT-PCR and generates a numeric RS of 0 to 100. The prospective-retrospective trials that validated this assay established that RS was both prognostic for recurrence of HR+ breast cancer and predictive of benefit from adjuvant chemotherapy, such that patients with high RS (originally defined as ≥31) benefited from chemotherapy, while patients with low RS (originally defined as <18) did not. However, chemotherapy benefit could not be excluded in patients with an intermediate RS of 18 to 30.

The TAILORx trial was designed to prospectively evaluate the 21-gene assay and to determine the benefit of adjuvant chemotherapy in HR+ HER2− node-negative breast cancer with intermediate RS. The trial enrolled 10,273 women under age 75 years with HR+ HER2− node-negative breast cancer who met National Comprehensive Cancer Network (NCCN) guidelines for recommendation of adjuvant chemotherapy (tumor 1.1–5 cm in size or 0.6–1.0 cm and intermediate to high histologic grade). Patients were assigned or randomized to chemotherapy based on results of the 21-gene assay. Patients with low RS (0–10) received ET alone; patients with high RS (≥26) received chemotherapy plus endocrine therapy (CET); and patients with a mid-range RS (11–25) were randomized to ET versus CET. The definitions of low, intermediate (or mid-range), and high RS were adjusted for the TAILORx trial in order to allow for the absence of HER2+ patients (who have high RS and benefit from chemotherapy and targeted HER2 therapy) in TAILORx, reflect the manner in which the assay is ordered in clinical practice, and minimize potential for undertreatment in the randomized arms. The primary end point was invasive disease-free survival (iDFS), defined as time from randomization to first event of local, regional, or distant recurrence, contralateral second primary invasive cancer, second primary nonbreast invasive cancer (excluding nonmelanoma skin cancer), or death without recurrence, and used a noninferiority design to determine whether patients with RS 11to25 derived benefit from adjuvant chemotherapy.

The initial results from TAILORx, published in 2015, focused on the low-risk patient cohort, which comprised 1626 women with RS 0 to 10. These patients had a 5-year iDFS of 93.8% (95% CI 92.4–94.9), rate of freedom from distant breast cancer recurrence of 99.3% (95% CI 98.7–99.6), rate of freedom from distant or locoregional breast cancer recurrence of 98.7% (95% CI 97.9–99.2), and overall survival (OS) of 98.0% (95% CI 97.1–98.6). Nine-year iDFS for this group was 84% (SE 1.3), and OS was 93.7% (SE 0.8). Thus TAILORx prospectively confirmed the excellent outcomes for HR+ HER2− node-negative breast cancer and low RS without chemotherapy.

Results from the randomized cohort, comprising 6711 patients with RS 11 to 25, showed that ET alone was noninferior to CET (HR for iDFS 1.08; 95% CI 0.94–1.24; P = 0.26). After 9 years of follow-up, iDFS was similar for patients receiving ET and CET (83.3% and 84.3%, respectively). Rates of freedom from distant recurrence (94.5% and 95.0%, respectively), freedom from distant and locoregional recurrence (92.2% and 92.9%), and OS (93.9% and 93.8%) were also similar in the two groups. Subgroup analyses showed an interaction between chemotherapy administration, age ≤0 years (or menopausal status), and RS, such that a small chemotherapy benefit was seen in women aged 50 years or younger with RS 16 to 26. Subsequent evaluation demonstrated that this benefit was concentrated among premenopausal women aged 45 to 50 years, suggesting that it was more associated with induction of menopause than with a direct antineoplastic effect of chemotherapy.

Further analysis of TAILORx evaluated the contribution of clinicopathologic risk factors (tumor size and grade) to the prognostic information provided by the RS. Clinical risk was prognostic for recurrence, but did not predict chemotherapy benefit. Women aged 50 years or younger with low clinical risk (as defined in the MINDACT trial ) and RS 16 to 20 had a very low 9-year distant recurrence risk with ET alone (4.6% + SE 1.5), and no benefit from chemotherapy. In contrast, young women with high clinical risk and RS 16 to 20, and women with RS 21 to 25 and both low and high clinical risk, had higher rates of distant recurrence, and did exhibit chemotherapy benefit (6.5% + SE 4.9% for RS 16–20 and high clinical risk; 6.4% + SE 4.9% for RS 21–25 and low clinical risk; 8.7% + SE 6.2% for RS 21–25 and high clinical risk). An online educational tool called RSClin was subsequently developed that integrates prognostic information provided by clinicopathologic features (including age, tumor size, and grade) and the prognostic and predictive information provided by the 21-gene RS. RSClin was developed using data derived from a patient-specific meta-analysis, validated for prognosis in an independent cohort, and provides both 10-year distant recurrence risk estimates with ET alone and estimated absolute chemotherapy benefit for women with node-negative breast cancer.

The 1389 patients enrolled in TAILORx with RS 26 to 100 had substantially poorer outcomes than those with RS 0 to 25, despite receipt of adjuvant chemotherapy, with 5-year iDFS of 87.6% (SE 1.0%), freedom from distant recurrence of 93% (SE 0.8%), and OS of 95.9% (SE 0.6%). No difference in outcomes was seen based on the chemotherapy regimen administered. The expected distant recurrence rate without chemotherapy for these patients was estimated based on the chemotherapy treatment effect observed in the HER2– cohort of the NSABP B20 trial, and was projected at 78.8% (SE 14.0%), suggesting a large absolute chemotherapy benefit associated with RS ≥26.

RxPONDER

While use of the 21-gene assay to guide adjuvant treatment in the node-negative setting was widely adopted even prior to the publication of TAILORx, NCCN guidelines endorsed adjuvant chemotherapy for all patients with nodal involvement as late as 2017. The ability of the 21-gene assay to predict chemotherapy benefit in patients with involvement of axillary nodes was prospectively evaluated in the RxPONDER trial, which randomized 5083 women with HR+, HER2– breast cancer, involvement of one to three axillary nodes, and RS of 0 to 25 to CET versus ET alone. The initial results from this trial showed that postmenopausal women did not benefit from chemotherapy, with 5-year iDFS of 91.9% for CET versus 91.6% for ET alone (HR = 0.97, P = 0.82), and 5-year OS of 96.2% versus 96.1% (HR = 0.96, P = 0.79). In contrast, premenopausal women receiving chemotherapy had improved iDFS (94.2% for CET vs. 89.0% for ET alone [HR = 0.54, P = 0.0004] and OS [98.6% vs. 97.3%, HR = 0.47, P = 0.032]). Improvement in iDFS was similar in women with RS 0 to 13 and 14 to 25. The majority of premenopausal women in this study did not receive ovarian suppression, suggesting that the chemotherapy benefit may have been predominantly due to induction of premature menopause, and that similar results may be achievable with optimization of ET.

Deescalation of Chemotherapy in HER2+ Disease

In 2005 the results of several randomized trials comparing adjuvant chemotherapy plus trastuzumab with adjuvant chemotherapy alone in HER2+ breast cancer were presented, demonstrating 50% reduction in risk of recurrence with addition of trastuzumab, and creating a new standard of care for HER2+ disease. However, the majority of patients enrolled in these studies had stage II or stage III disease. The chemotherapy backbones used in these trials, either doxorubicin, cyclophosphamide and paclitaxel, or docetaxel and carboplatin, are associated with substantial acute toxicity, as well as long-term and late side effects such as neuropathy and anthracycline-associated cardiotoxicity. While stage I HER2+ tumors were underrepresented in the adjuvant therapy trials, and were expected to derive less absolute benefit from adjuvant therapy, they still have a poorer prognosis than HER2− disease in the absence of adjuvant cytotoxic therapy.

In 2015 Tolaney and colleagues published the results of the APT trial ( Table 3.1 ), a phase II study evaluating a less toxic adjuvant regimen consisting of 12 weeks of weekly paclitaxel plus trastuzumab in women with early HER2+ breast cancer. The trial enrolled 410 women with HER2+ breast cancer measuring 3 cm or less. Originally only patients with node-negative disease were eligible, but the trial was amended to allow patients with micrometastasis to one axillary node. Only 1.5% of patients had nodal involvement. At median follow-up of 4 years, 3-year rate of survival free from invasive disease was 98.7% (95% CI 97.6–99.8). Only two distant breast cancer relapses were seen. Updated analysis after a median follow-up of 6.5 years showed a 7-year DFS rate of 93.3% (95% CI 90.4–96.2) with four (1.0%) distant recurrences, and 7-year OS rate of 95% (95% CI 92.4–97.7). Although not a randomized trial, the excellent outcomes observed and good tolerability of this regimen have led to its wide adoption and endorsement by NCCN guidelines.

Optimization of Neoadjuvant Therapy

Patients with locoregionally advanced breast cancer that is operable may receive systemic therapy either preoperatively or postoperatively. Randomized trials have demonstrated similar outcomes when chemotherapy is administered in the neoadjuvant or the adjuvant setting in patients with localized operable breast cancer. However, for those with regionally advanced disease, neoadjuvant chemotherapy, used alone in HER2– disease or in combination with anti-HER2 therapy in HER2+ disease, may facilitate breast conservation and reduce the extent of axillary surgery. In addition, achievement of pathologic complete response (pCR) to neoadjuvant therapy is associated with improved breast cancer outcomes. A meta-analysis of 12 trials involving 11,955 patients established the definition of pCR as absence of invasive carcinoma in both breast and axillary nodes (ypT0N0 or ypTisN0), and demonstrated that the association of pCR with improved event-free survival (EFS) and OS is strongest for triple-negative disease and for HER2+ disease treated with trastuzumab-containing regimens. The US Food and Drug Administration (FDA) determined that pCR could be used as a surrogate end point for approval of new drugs, and that a randomized trial that added a novel agent to a standard neoadjuvant backbone could serve as the basis for such approval. In addition, lack of pCR after optimal neoadjuvant systemic therapy may serve as a pharmacodynamic biomarker for identifying patients at higher recurrence risk and thus more likely to benefit from additional adjuvant systemic therapy.

Over the past decade, several important trials have focused on optimization of neoadjuvant therapy to increase pCR rate, as well as intensification of adjuvant treatment for patients with suboptimal response to neoadjuvant therapy (summarized in Table 3.2 ). The most innovative of these studies is the ongoing I-SPY2 trial, an open-label adaptive phase II trial evaluating addition of novel agents to standard neoadjuvant chemotherapy in patients with stage II or III breast cancer. The trial classifies breast cancer by molecular signatures based on HR and HER2 status and results of MammaPrint assay, and uses results to adaptively randomize patients to either standard chemotherapy with 12 weeks of paclitaxel (plus anti-HER2 therapy for patients with HER2+ disease) followed by four cycles of doxorubicin and cyclophosphamide or to one of several experimental arms. Agents “graduate” from I-SPY2 when they demonstrate sufficient activity in any biomarker signature to have an 85% probability of success in a phase III trial. Multiple therapies have been evaluated in this trial, including pertuzumab and pembrolizumab, which are now FDA-approved in the neoadjuvant setting.

Table 3.2

Randomized Phase II or III Studies Evaluating Neoadjuvant and Post-Neoadjuvant Therapy in Locally Advanced Breast Cancer

Trial	n	Design	Population	Randomization	Primary End Point	Findings
NeoSphere	417	Phase II	T > 2 cm, N0, N+, HER2+	Neoadjuvant DH vs. DP vs. DHP vs. HP	pCR	Increase in pCR with DHP compared with DH 16.8% pCR rate for HP, despite no cytotoxic chemotherapy
TRYPHAENA	225	Phase II	T > 2 cm, N0, N+, HER2+	Neoadjuvant FEC-HP→DHP vs. FEC ->DHP vs. DCHP	Cardiac toxicity	Low rates of cardiotoxicity in all arms pCR rates 57% to 66%
KEYNOTE-522	1174	Phase III	Stage II or III, TN	Neoadjuvant TC-pembrolizumab→AC/EC vs. TC-placebo→AC/EC; adjuvant pembrolizumab vs. placebo	pCR; EFS	Increased pCR and EFS with pembrolizumab
IMPASSION-031	455		Stage II or III TN	Neoadjuvant nab-paclitaxel-atezolizumab→ddAC vs. nab-paclitaxel-placebo→ddAC; adjuvant atezolizumab vs. placebo	pCR in ITT pts; pCR in PD-L1+	Increased pCR with atezolizumab
KATHERINE	1486	Phase III	T > 1 cm, N0, N+, HER2+, no pCR to neoadjuvant therapy	Postoperative T-DM1 vs. trastuzumab	iDFS	Improved iDFS with T-DM1
CREATE-X	910	Phase III	Stage I to IIIB, HER2−, HR+ and HR−, no pCR to neoadjuvant therapy	Postoperative capecitabine vs. no chemotherapy (endocrine therapy for HR+)	DFS	Improved DFS and OS with capecitabine Benefit confined to TN

AC , Doxorubicin and cyclophosphamide; ddAC , dose-dense doxorubicin and cyclophosphamide; DCHP , docetaxel, carboplatin, trastuzumab, and pertuzumab; DFS , disease-free survival; DH , docetaxel and trastuzumab; DHP , docetaxel, trastuzumab, and pertuzumab; DP , docetaxel and pertuzumab; EC , epirubicin and cyclophosphamide; EFS , event-free survival; FEC->DCP , 5-fluorouracil, epirubicin, and cyclophosphamide, followed by DHP; FEC-HP->DHP , 5-fluorouracil, epirubicin, cyclophosphamide, trastuzumab, and pertuzumab, followed by DHP; HP , trastuzumab and pertuzumab; HR+ , hormone receptor–positive; HR− , hormone receptor–negative; HER2+ , HER2-positive; iDFS , invasive disease-free survival; ITT , intent to treat; N0 , node negative; N+ , node positive; OS , overall survival; pCR , pathologic complete response; PD-L1+ , programmed death ligand 1-positive; pts , patients; T , tumor size; TC , paclitaxel and carboplatin; T-DM1 , ado-trastuzumab emtansine; TN , triple negative.

Studies Evaluating Neoadjuvant Dual Anti-HER2 Therapy in HER2-Positive Breast Cancer

Two randomized phase II trials, NeoSphere and TRYPHAENA, evaluated the addition of pertuzumab to neoadjuvant therapy in HER2+ breast cancer. In NeoSphere, 417 women were randomized to four cycles of neoadjuvant therapy including docetaxel plus trastuzumab; docetaxel plus pertuzumab; all three agents; or trastuzumab and pertuzumab (HP) without cytotoxic chemotherapy. The primary end point was pCR in the breast. Patients receiving docetaxel plus HP had significantly improved pCR rates over patients treated with trastuzumab plus docetaxel (45.8% [95% CI 36.1–55.7] vs. 29.0% [20.6–38.5]; P = 0.0141). Interestingly, 16.8% of patients receiving HP alone achieved pCR, despite receiving no cytotoxic chemotherapy. Although recurrence rates were similar, the trial was underpowered to detect difference in clinical outcomes other than pCR rates.

The TRYPHAENA study was designed to assess the cardiotoxicity of HP added to anthracycline-containing and non–anthracycline-containing chemotherapy regimens. Two hundred and twenty-five patients were randomized to six cycles of neoadjuvant therapy, including three cycles of 5-fluorouracil, epirubicin, and cyclophosphamide (FEC) plus HP followed by three cycles of docetaxel plus HP (FEC-HP→DHP); three cycles of FEC without HP followed by three cycles of docetaxel plus HP (FEC→DHP); or six cycles of docetaxel, carboplatin, and HP (DCHP). The primary end points were rates of symptomatic left ventricular systolic dysfunction (LVSD) and declines in left ventricular ejection fraction (LVEF) by ≥10% to a level below 50% during neoadjuvant therapy. Low cardiotoxicity rates were seen in all three arms. Symptomatic LVSD during neoadjuvant therapy was seen in only two patients (2.7%), both of whom were enrolled in the FEC→DHP arm, and one of whom developed LVSD prior to initiating anti-HER2 therapy. LVEF declines of >10% to below 50% during neoadjuvant therapy were seen in four patients (5.6%) receiving FEC-HP→DHP, three patients (4.0%) receiving FEC→DHP, and two patients (2.6%) receiving DCHP. All cardiac toxicities were reversible. The study was not designed to compare efficacy of the three arms, but high rates of pCR (57%–66%) were seen. Based on results of NeoSphere and TRYPHAENA, the FDA granted accelerated approval to neoadjuvant pertuzumab in combination with trastuzumab and docetaxel in 2013.

You're Reading a Preview

Become a Clinical Tree membership for Full access and enjoy Unlimited articles

Become membership

If you are a member. Log in here