No definite preferred standard of care exists for adjuvant chemotherapy (ACT) in patients with node-negative, HER2-positive breast cancers measuring <2 cm.1 Prior to the advent of HER2-directed therapy, HER2 overexpression was a marker of poor prognosis, and is currently found in 15% to 20% of all invasive breast cancers. Approximately 10% of small, node-negative invasive breast cancers will also overexpress HER2.2–4 Trastuzumab is a monoclonal antibody that targets an extracellular domain of the HER2 oncogene. Trastuzumab with ACT has been shown to improve disease-free survival (DFS) and overall survival (OS) in previous trials, and has become the standard of care in patients with high-risk node-negative or node-positive breast cancer.5–7 However, these landmark trials often excluded or underrepresented patients with node-negative breast cancer measuring <2 cm.
For patients with node-negative, HER2-positive breast cancers ≤1 cm, NCCN used expert opinion (category 2B) to recommend consideration of trastuzumab-containing ACT regimens, and category 1 evidence to support its use in patients with tumors >1 cm.8 Data supporting use of chemotherapy for node-negative, HER2-positive breast cancers ≤1 cm are limited by the small, single-armed, or observational nature of the studies.9–12 Patients with node-negative tumors that are 1 to 2 cm are underrepresented in the prospective randomized trials, which showed improvements in relative risk reduction but less clear absolute benefits.7,13,14 Despite the lack of robust evidence, a significant increase in the use of trastuzumab-based chemotherapy has occurred in patients with node-negative, HER2-positive breast cancers ≤1 cm in the past decade.15 Because trastuzumab-based chemotherapy has both short- and long-term adverse effects, the question should be asked whether there is a tumor size below which treatment has little value or is even detrimental in terms of overall outcome. Therefore, we used the National Cancer Database (NCDB) to evaluate the use and impact of ACT on OS in patients with node-negative, HER2-positive breast cancers <2 cm.
Methods
Data Source
Access was granted to the NCDB Participant Use Data File (PUF) for patients diagnosed with breast cancer from 2004 through 2014. Patients were identified based on International Classification of Diseases for Oncology, Third Edition (ICD-O-3) site codes C50.0 through C50.9.16 The NCDB is a nationwide, facility-based, comprehensive, clinical surveillance resource oncology data set, started jointly in 1989 by the Commission on Cancer (CoC) of the American College of Surgeons and the American Cancer Society, that integrates hospital registry data collected in >1,500 CoC-accredited facilities. The data represent approximately 70% of all newly diagnosed cancer cases nationwide annually. The NCDB PUF is a HIPAA-compliant data file. Institutional Review Board approval was not required for this study because the PUF is deidentified.
Study Population
After obtaining permission from the NCDB, we queried the 2014 NCDB PUF for patients who were diagnosed with histologically confirmed invasive breast cancer from 2010 to 2013. We limited our analysis to 2010 through 2013 because the NCDB began requiring HER2 reporting in 2010. Vital status information is not included for patients diagnosed in 2014 because of the limited follow-up for these patients. Follow-up ranged from 6 years for patients diagnosed in 2010 to 3 years for those diagnosed in 2013. The PUF data do not include cause of death information, and therefore cause-specific survival cannot be calculated. In our analysis, we included adults aged ≥18 and ≤70 years to maximize the number of patients treated with chemotherapy and minimize the non–cancer-related deaths. We selected female patients with node-negative, HER2-positive invasive breast cancer <2 cm. Patients had to have undergone definitive surgery (lumpectomy with radiation or mastectomy) with negative margins. We excluded those who were HER2-negative or had unknown HER2 status, whose survival time was missing, and who did not receive definitive surgery. Figure 1 depicts the CONSORT diagram.
Stratification of patients into treated and untreated groups was complicated by the fact that the NCDB does not provide drug-specific data. For patients diagnosed between 2010 and 2012, some biologic agents, including trastuzumab and pertuzumab, were classified as chemotherapy. In 2013, the NCDB announced a change in classification for trastuzumab and pertuzumab, and these agents were reclassified as immunotherapy for cases diagnosed starting January 1, 2013. This change impacted our data set, and we opted to include patients who received chemotherapy with or without explicitly noted immunotherapy treatment in our treatment group. For the purposes of this study, the term “chemotherapy” will be used to include systemic chemotherapy, with anti-HER2 therapy including trastuzumab and pertuzumab. The nontreatment group included only patients without explicitly noted immunotherapy who did not receive chemotherapy. The small group of patients recorded as receiving immunotherapy without chemotherapy were excluded.
Statistical Methods and Analysis
We assessed potential associations between sociodemographic, stage, estrogen receptor (ER)/progesterone receptor (PR) status, and tumor and treatment characteristics using Pearson chi-square or Fisher exact tests. Survival analysis using Kaplan-Meier curves
CONSORT diagram of the study population.
Abbreviation: NCDB, National Cancer Database.
Citation: Journal of the National Comprehensive Cancer Network J Natl Compr Canc Netw 16, 11; 10.6004/jnccn.2018.7058
Results
Patient Characteristics
A total of 21,148 patients identified in the database between 2010 and 2013 met inclusion criteria. Baseline patient characteristics are summarized in Table 1. Most were white (82.2%), were aged ≥50 years (71.3%), had good performance status (87.1% with Charlson/Deyo score of 0), had some form of insurance (96.9%), had no other prior malignancy (87.9%), had ductal carcinoma histology (92.9%), and had T1c stage disease at diagnosis (48.2%). Between 2010 and 2013, marked shifts occurred in the number of patients classified as receiving ACT alone compared with ACT combined with immunotherapy (Table 2). Associations between stage and clinicodemographic characteristics are summarized in Table 3.
Association Between Treatment Receipt and Tumor and Sociodemographic Characteristics
An increased use of ACT occurred during the study period. The annual treatment rates from 2010 to 2013 were 71.5%, 72.4%, 73.3%, and 74.4%, respectively (trend test, P<.0001). Increasing rates of ACT were primarily driven by increased treatment rates for T1b and T1c tumors (trend test, P<.0001 and P<.0001, respectively), whereas treatment rates for T1a tumors did not change significantly (trend test, P=.64) and rates decreased slightly in T1mi tumors (trend test, P=.03). Practice patterns in use of ACT reflected NCCN recommendations, in that use increased from T1mi to T1c (26.1%, 46.3%, 77.5%, 88.2%; trend test, P<.0001) and from well-differentiated to poorly/undifferentiated tumors (48.4% to 82.5%; P<.0001). Younger patients were more likely to receive treatment (87.0% for <40 years of age vs 67.4% for 60–70 years of age; P<.0001). ER/PR status was not associated with receipt of chemotherapy. A total of 82% of all patients with ER-positive breast cancer received endocrine therapy. Patients with malignant cells in regional lymph nodes that measured <0.2 mm (N0i+ or N0mol+) were more likely to receive ACT than those without malignant cells in regional lymph nodes N0 (including N0i− or N0mol−).
Baseline Demographic and Clinical Characteristics
Patients residing in areas with low educational attainment (≥21% of the population without a high school degree), with poor performance status (Charlson/Deyo score of ≥2), or a history of previous cancer were less likely to receive ACT. Academic/Research
Reporting in the NCDB
Overall Survival
In the entire cohort (n=21,148), ACT was associated with decreased 5-year OS rates in patients with T1mi (n=1,452; 89.6% [95% CI, 83.7%–93.4%] vs 98.1% [96.6%–98.9%]; log-rank P<.0001) and
Associations Between Stage and Clinicodemographic Characteristics
In our 1:1 propensity-matched cohort model (n=8,222), ACT was associated with a decreased 5-year OS rate in patients with T1mi tumors (n=626; 89.1% [95% CI, 81.8%–93.5%] vs 99.1% [96.6%–99.8%]; P=.0006) (Figure 2A), and no effect on 5-year OS was seen in patients with T1a tumors (n=2,901, 95.4% [93.2%–96.9%] vs 96.9% [94.1%–98.3%]; P=.059) (Figure 2B). It is important to note, however, that the overall number of events in patients with T1mi tumors was small, with 18 observed events in the ACT group and 2 observed events in the group that did not receive ACT. Patients with T1b (n=2,340) and T1c tumors (n=2,355) had improved 5-year OS rates with treatment (97.1% [95.1%–98.4%] vs 92.3% [88.5%–94.9%]; P=.0016 and 95.9% [93.5%–97.5%] vs 91.5% [88.4%–93.9%]; P=.0002, respectively) (Figure 2C, D, respectively).
Among the entire cohort, OS was poorer in patients with ER/PR-negative tumors compared with ER/PR-positive tumors (hazard ratio [HR], 1.51 [95% CI, 1.23–1.85]; P<.0001), regardless of tumor grade and/or chemotherapy treatment (poorly/undifferentiated vs moderately/well-differentiated: HR, 1.12 [0.92–1.40]; P=.25).
Discussion
To our knowledge, this is the largest registry-based study in the trastuzumab era examining the use and impact of ACT on OS in patients with node-negative, HER2-positive breast cancer measuring <2 cm. Our data support the use of ACT, showing an association with improved OS for patients with T1b and T1c node-negative, HER2-positive breast cancer. An association with decreased OS was observed for ACT in patients with T1mi, and no benefit was observed in patients with T1a node-negative, HER2-positive breast cancer. We observed an increasing use of chemotherapy during the study period.
Node-negative, HER2-positive breast cancer <2 cm is a challenging tumor type in which to conduct
Stage-specific overall survival after treatment with versus without adjuvant chemotherapy for (A) T1mi, (B) T1a, (C) T1b, and (D) T1c tumors.
Citation: Journal of the National Comprehensive Cancer Network J Natl Compr Canc Netw 16, 11; 10.6004/jnccn.2018.7058
Adding to the complex medical decision-making is the poor characterization of risk in patients with node-negative, HER2-positive breast cancer <2 cm. In a smaller Finnish registry (N=500; T1a/b=171, T1c=329), the risk of recurrence was characterized as high, with only 72% DFS at 9 years.10 However, larger registry and national database reports have shown significantly lower rates of recurrence and favorable OS. In the MD Anderson series, 5-year DFS with pT1a and pT1b tumors was 92% for both groups, without ACT and trastuzumab.4 In a pre-trastuzumab era report, a SEER registry–based analysis of 51,246 patients with T1a or T1b, node-negative, HER2-positive breast cancer had a breast cancer–specific mortality of 4%.21 Our NCDB assessment supports relatively high 5-year OS rates among patients with T1mi and T1a disease who did not receive ACT in both our overall and propensity-matched cohorts (T1mi: 98.1% [95% CI, 96.6%–98.9%] and 99.1% [96.6%–99.8%], respectively; T1a: 96.5% [94.6%–97.7%] and 96.9% [94.1%–98.3%], respectively). However, patients with T1b and T1c disease had a poorer outcome without ACT in both our overall and propensity-matched cohorts (T1b: 92.3% [88.7%–94.8%] and 92.3% [88.5%–94.9%], respectively; T1c: 91.6% [88.5%–93.9%] and 91.5% [88.4%–93.9%]). We could not assess DFS in the NCDB.
Risk stratification of T1a and T1b tumors based on clinicopathologic features, including hormone receptor status, tumor grade, and Ki-67 index, has been attempted in previous studies. ER status in node-negative, HER2-positive tumors <2 cm has not been associated with outcome in several prior trials.4,10,22 To our knowledge, the only prior trial to show an effect based on hormone receptor status was the EIO series,2 reporting a nonintuitive reduced DFS in patients with node-negative, ER-positive, HER2-positive tumors <2 cm (HR, 5.2 vs 2.4 for the whole HER2-positive population). Our analysis supports hormone receptor status as a predictive risk factor in node-negative breast cancer <2 cm, because patients with ER/PR-negative tumors had poorer survival overall than those with ER- or PR-positivity (HR, 1.51 [95% CI, 1.23–1.85]; P<.0001), regardless of tumor grade (poorly/undifferentiated vs moderately/well-differentiated: HR, 1.12 [0.92–1.40]; P=.25). In our entire cohort, receipt of endocrine therapy was also associated with improved OS. Ki-67 index and tumor grade have also been evaluated as clinicopathologic risk factors in HER2-positive breast cancer.10,23 Although we could not report on Ki-67 because it is not a reported variable in the NCDB, tumor grade was assessed and found not to be associated with OS in our patient population.
Our 5-year OS data in patients who received ACT in both the overall and propensity-matched cohorts (T1b: 96.8% [95% CI, 95.6%–97.7%] and 97.1% [95.1%–98.4%]; T1c: 95.8% [94.9%–96.5%] and 95.9% [93.5%–97.5%]) are very similar to those from a phase II, prospective, nonrandomized study of weekly adjuvant paclitaxel in combination with trastuzumab for 12 weeks, followed by maintenance trastuzumab for 1 year in patients with node-negative, HER2-positive breast cancer ≤3 cm.11 In that study, 201 of the 406 participants had pT1a/bN0 breast cancer, and the 3-year DFS rate was 98.7%, a proportion significantly higher than the historical data from the MD Anderson series,4 among others. An update of this data set was reported with a median follow-up of 6.5 years, showing that the 7-year DFS was 93.3% (95% CI, 90.4%–96.2%) and 7-year OS was 95.0% (95% CI, 92.4%–97.7%).24 More aggressive chemotherapy regimens may be unwarranted in node-negative, HER2-positive breast cancer <2 cm.
We did observe an association between decreased survival and receipt of ACT in patients with T1mi breast cancer in the propensity-matched cohort and in those with T1mi and T1a disease in the unmatched cohort. However, due to uncontrolled variables in the unmatched cohort, small sample size in the matched cohort (N=379 with mature 5-year OS data), and low number of events overall, we were not able to draw significant conclusions.
We showed an increasing use of ACT across all years in our overall cohort and in patients with T1b and T1c tumors. This finding is consistent with prior reports of increasing ACT use.15 We observed relatively high use of ACT in patients with T1mi and T1a breast cancer (26.1% and 46.3%, respectively), and suggest selective use of chemotherapy in this subset given the lack of observed benefit in our report, potential risk of harm observed, and risk of toxicity.
Strengths of this study include the sample size afforded by the NCDB, allowing us to capture most patients in the United States with node-negative, HER2-positive, chemotherapy-eligible breast cancer <2 cm. We were able to construct a propensity-matched cohort and adjust for the known confounding patient, tumor, and facility factors that affect receipt of chemotherapy and OS.
Our study has limitations, including its retrospective design and relatively short follow-up. In addition, unobserved confounding factors can limit interpretation of study outcomes derived from observational data. Some misclassification and treatment underreporting are unavoidable in a large registry-based data set like the NCDB. The NCDB does not provide drug-specific information regarding the chemotherapy administered, DFS, disease-specific survival, and local recurrence. There is a known under-ascertainment of immunotherapy in the NCDB as reported compared with linked databases such as the SEER-Medicare database.25 Additionally, the change in coding of biologic agents in 2013 of trastuzumab and pertuzumab impacted our data set, limiting our ability to definitively identify patients receiving chemotherapy that included HER2-directed therapy. Due to these limitations, we could not assess the chemotherapy regimen recommended, the inclusion of HER2-directed therapy, breast cancer recurrences, or progression-free survival. Our OS data are limited to 5 years given the short duration of HER2 data in the NCDB, with reporting starting in 2010.
Conclusions
ACT was found to be associated with improved OS in T1b and T1c breast cancer and no benefit in T1mi and T1a tumors in a nationally representative cohort of chemotherapy-eligible women with node-negative, HER2-positive, breast cancer <2 cm in the trastuzumab era. During the study period, an increasing use of chemotherapy was observed. We support chemotherapy use in T1b and T1c node-negative, HER2-positive breast cancer <2 cm, but urge more judicious use in patients with T1mi and T1a disease. We caution that treatment recommendations need to be individualized in this patient population. Future investigations are warranted using this data set, with 10- and 15-year follow-up data when mature, as future randomized clinical trials are unlikely.
References
- 1.↑
Edge SB, Byrd DR, Compton CC et al.. AJCC Cancer Staging Handbook, 7th ed. Springer–Verlag New York: New York, NY; 2010.
- 2.↑
Curigliano G, Viale G, Bagnardi V et al.. Clinical relevance of HER2 overexpression/amplification in patients with small tumor size and node-negative breast cancer. J Clin Oncol 2009;27:5693–5699.
- 3.
Chia S, Norris B, Speers C et al.. Human epidermal growth factor receptor 2 overexpression as a prognostic factor in a large tissue microarray series of node-negative breast cancers. J Clin Oncol 2008;26:5697–5704.
- 4.↑
Gonzalez-Angulo AM, Litton JK, Broglio KR et al.. High risk of recurrence for patients with breast cancer who have human epidermal growth factor receptor 2–positive, node-negative tumors 1 cm or smaller. J Clin Oncol 2009;27:5700–5706.
- 5.↑
Romond EH, Perez EA, Bryant J et al.. Trastuzumab plus adjuvant chemotherapy for operable HER2-positive breast cancer. N Engl J Med 2005;353:1673–1684.
- 6.↑
Piccart-Gebhart MJ, Procter M, Leyland-Jones B et al.. Trastuzumab after adjuvant chemotherapy in HER2-positive breast cancer. N Engl J Med 2005;353:1659–1672.
- 7.↑
Slamon D, Eiermann W, Robert N et al.. Adjuvant trastuzumab in HER2-positive breast cancer. N Engl J Med 2011;365:1273–1283.
- 9.↑
Fehrenbacher L, Capra AM, Quesenberry J et al.. Distant invasive breast cancer recurrence risk in human epidermal growth factor receptor 2-positive T1a and T1b node-negative localized breast cancer diagnosed from 2000 to 2006: a cohort from an integrated health care delivery system. J Clin Oncol 2014;32:2151–2158.
- 10.↑
Joensuu H, Isola J, Lundin M et al.. Amplification of erbB2 and erbB2 expression are superior to estrogen receptor status as risk factors for distant recurrence in pT1N0M0 breast cancer: a nationwide population-based study. Clin Cancer Res 2003;9:923–930.
- 11.↑
Tolaney SM, Barry WT, Dang CT et al.. Adjuvant paclitaxel and trastuzumab for node-negative, HER2-positive breast cancer. N Engl J Med 2015;372:134–141.
- 12.↑
Jones SE, Collea R, Paul D et al.. Adjuvant docetaxel and cyclophosphamide plus trastuzumab in patients with HER2-amplified early stage breast cancer: a single-group, open-label, phase 2 study. Lancet Oncol 2013;14:1121–1128.
- 13.↑
Cameron D, Piccart-Gebhart MJ, Gelber RD et al.. 11 years' follow-up of trastuzumab after adjuvant chemotherapy in HER2-positive early breast cancer: final analysis of the HERceptin Adjuvant (HERA) trial. Lancet 2017;389:1195–1205.
- 14.↑
Joensuu H, Kellokumpu-Lehtinen PL, Bono P et al.. Adjuvant docetaxel or vinorelbine with or without trastuzumab for breast cancer. N Engl J Med 2006;354:809–820.
- 15.↑
Vaz-Luis I, Ottesen RA, Hughes ME et al.. Outcomes by tumor subtype and treatment pattern in women with small, node-negative breast cancer: a multi-institutional study. J Clin Oncol 2014;32:2142–2150.
- 16.↑
National Cancer Data Base Participant Use Data File (PUF) Data Dictionary. Version PUF 2013. Available at: https://www.facs.org/~/media/files/quality%20programs/cancer/ncdb/puf%20data%20dictionary%20version%20puf%202014.ashx. Updated 2015. Accessed April 7, 2016.
- 17.↑
Smith I, Procter M, Gelber RD et al.. 2-year follow-up of trastuzumab after adjuvant chemotherapy in HER2-positive breast cancer: a randomised controlled trial. Lancet 2007;369:29–36.
- 18.
Spielmann M, Roché H, Delozier T et al.. Trastuzumab for patients with axillary-node–positive breast cancer: results of the FNCLCC-PACS 04 trial. J Clin Oncol 2009;27:6129–6134.
- 19.↑
Perez EA, Romond EH, Suman VJ et al.. Trastuzumab plus adjuvant chemotherapy for human epidermal growth factor receptor 2-positive breast cancer: planned joint analysis of overall survival from NSABP B-31 and NCCTG N9831. J Clin Oncol 2014;32:3744–3752.
- 20.↑
O'Sullivan CC, Bradbury I, Campbell C et al.. Efficacy of adjuvant trastuzumab for patients with human epidermal growth factor receptor 2-positive early breast cancer and tumors ≤2cm: a meta-analysis of the randomized trastuzumab trials. J Clin Oncol 2015;33:2600–2608.
- 21.↑
Hanrahan EO, Gonzalez-Angulo AM, Giordano SH et al.. Overall survival and cause-specific mortality of patients with stage T1a,bN0M0 breast carcinoma. J Clin Oncol 2007;25:4952–4960.
- 22.↑
Tovey SM, Brown S, Doughty JC et al.. Poor survival outcomes in HER2 positive breast cancer patients with low grade, node negative tumours. Br J Cancer 2009;100:680–683.
- 23.↑
Colleoni M, Rotmensz N, Peruzzotti G et al.. Minimal and small size invasive breast cancer with no axillary lymph node involvement: the need for tailored adjuvant therapies. Ann Oncol 2004;15:1633–1639.
- 24.↑
Tolaney SM, Barry WT, Guo H et al.. Seven-year (yr) follow-up of adjuvant paclitaxel (T) and trastuzumab (H) (APT trial) for node-negative, HER2-positive breast cancer (BC) [abstract]. J Clin Oncol 2017;35(Suppl):Abstract 511.
