One year of adjuvant trastuzumab remains standard of care
Harold J. Burstein, MD, PhD
Approximately 20% to 25% of human breast cancers are marked by an acquired alteration or over-expression in the HER2 gene. While HER2-positive tumors can be aggressive, researchers and clinicians have developed an armamentarium of several drugs, as well as combination therapy approaches, to attack this disease.1
One of those approaches includes the addition of trastuzumab (Herceptin, Genentech) to chemotherapy for 1 year, a strategy that has been largely effective. However, questions around the use of a shorter trastuzumab course in this setting have appeared frequently enough in clinical trials to warrant further discussion.
“The important thing to remember is that in 2002, when developing adjuvant trials of trastuzumab, researchers arbitrarily chose 1 year as the duration,” Harold J. Burstein, MD, PhD, associate professor of medicine at Harvard Medical School, said in an interview. “Specifically, that duration was picked for the HERA trial.”
Results of the HERA study showed that 2 years of adjuvant trastuzumab was no better than 1 year, and so 1 year became the standard of care, according to Burstein.
“The question since then has become: Do you need that much, or can you do with less?” he said.
Five trials have investigated that question: PERSEPHONE, PHARE, Short-HER, HORG and SOLD. Burstein, who co-authored an editorial in Annals of Oncology with Otto Metzger Filho, MD, of Dana-Farber Cancer Institute on this topic, offered a broad conclusion about these trials.2
“None of these trials showed that the shorter duration of treatment was non-inferior to 1 year,” he said.
However, Burstein acknowledged that the differences in efficacy were often statistically negligible. Additionally, when adverse outcomes like cardiac toxicity are factored into the equation, the issue may not be so clear-cut. Another complication is resource utilization of this prolonged and costly therapy.
In the landmark phase 3, non-inferiority PERSEPHONE trial, Earl and colleagues compared 6 versus 12 months of adjuvant trastuzumab in a cohort of 4,089 patients. At a median follow-up duration of 4.9 years, 500 DFS events and 319 deaths had occurred. The 4-year DFS rate was 89% (95% CI, 88% to 91%) in both arms.3
The researchers outlined an HR non-inferiority limit of 1.29. The observed HR was 1.05 (95% CI, 0.88 to 1.25), which established the non-inferiority of the 6-month regimen (1-sided P = .01).3
Other findings showed “congruent results” for OS, as well as the pre-planned DFS and OS landmark analyses following 6 months of trastuzumab, although heterogeneity was noted in some stratification variables, according to the researchers. Cardiac events leading to discontinuation occurred in 4% of patients treated for 6 months and in 8% of those treated for 12 months (P < .0001).3
“PERSEPHONE has demonstrated 6 months of trastuzumab as non-inferior to 12 months” with a 3% non-inferiority margin, the researchers concluded. “Given cardiac and other toxicities during months 7-12 of treatment, our results would support a reduction of standard trastuzumab duration to 6 months.”3
PERSEPHONE was the only trial to demonstrate that the shorter duration of trastuzumab appeared to be better than the longer regimen, according to Burstein. However, he also noted that patients in PERSEPHONE were treated with sequential chemotherapy and trastuzumab, which he believes may have contributed to the outcome.
“The issue with this trial is that many of the women were not treated with the standard-of-care of concurrent chemotherapy and trastuzumab,” he continued.
In a cost-effectiveness analysis of the PERSEPHONE trial, Hulme and colleagues wrote that the annual per-patient cost of trastuzumab exceeded £30,000 ($39,766 U.S.). These researchers looked at cost and quality of life parameters between 6 and 24 months after treatment began.4
Their analysis included findings for 4,009 patients who were disease-free at 6 months, which was further broken down into one group of 2,000 patients who were disease-free after 6 months and another group of 2,009 patients who were disease-free at 12 months. Results showed that 6 months of therapy yielded an average cost of £2,538.64 ($3,365.11 U.S.; 95% CI, £2,383.38 to £2,700.72) compared with £12,333.83 ($16,349.18 U.S.; 95% CI, £12,098.58 to £12,562.27) for 12 months. The researchers calculated an average per-patient cost savings of £9,793.25 ($12,981.50 U.S.; 95% CI, £9,515.86 to £10,071.64). Trastuzumab treatment and administration comprised £9,699.58 ($12,857.34 U.S.; 95% CI, £9,436.20 to £9,954.67) of this savings. The researchers concluded that the probability of 6 months being cost-effective is 100%.4
“It is worth considering that, in the kind of financially compromised situations you might find in the developing world, if you can treat two patients for 6 months rather than one patient for 12 months, this might be an attractive choice,” Burstein said. “It’s a matter of value. But, in the U.S., this one- versus two-patient Sophie’s choice model is not the way clinical medicine is set up.”
In the PHARE trial, Pivot and colleagues investigated 6 versus 12 months of trastuzumab in a cohort of 3,380 women with early HER2-positive breast cancer. The researchers observed no axillary node involvement in about half of the cases and approximately 40% of patients had ER-negative disease.5
The primary objective was non-inferiority of 6 versus 12 months in the intent-to-treat population in regard to DFS. The pre-specified hazard ratio margin for DFS was 1.15. Secondary endpoints included OS and metastasis-free survival.5
After a median follow-up period of 7.5 years, 704 DFS events had occurred. Between the 6-month and 12-month arms, the adjusted HR for DFS events favored the longer course of therapy (HR = 1.08; 95% CI, 0.93 to 1.25). No heterogeneity for treatment effect was demonstrated, according to the researchers, and no substantial difference for trastuzumab duration effects was observed in any subgroup. Adjusted hazard ratios were comparable for both OS (HR = 1.13; 95% CI, 0.92 to 1.39) and metastasis-free survival (HR = 1.15; 95% CI, 0.96 to 1.37).5
“The choice of the non-inferiority margin will remain inherently a subject of controversy, especially in the context of oncology trials where the primary outcome is survival and the least additional death could be considered unacceptable, questioning the very feasibility of such trials,” the researchers concluded. “Nevertheless, PHARE failed to show that 6 months of adjuvant trastuzumab was non-inferior to 12 months. The standard of care should remain 12 months of adjuvant trastuzumab.”5
A “statistical game” emerges when you begin examining the other trials, according to Burstein.
“PHARE showed a 2% to 3% difference between the two groups,” he told Healio. “The numerical differences are really small.”
In the SOLD study, Joensuu and colleagues enrolled 2,174 women with histologically confirmed HER2-positive breast cancer with either regional node-positive or node-negative disease. Patients received 3 cycles of docetaxel 3 times per week plus trastuzumab for 9 weeks, followed by 3 cycles of fluorouracil, epirubicin and cyclophosphamide. Patients in the 9-week group did not receive trastuzumab after that.6
Over a median follow-up period of 5.2 years (interquartile range, 3.8 to 6.7 years), the researchers failed to demonstrate non-inferiority of the 9-week regimen in terms of DFS (HR = 1.39; 2-sided 90% CI, 1.12 to 1.72). The two regimens were also comparable in terms of distant DFS and OS.6
Cardiac failure was reported in 2% of patients (n = 21) treated for 9 weeks and 3% (n = 36) of those treated for 1 year. Left ventricle ejection fraction was also maintained better in patients treated with 9 weeks of trastuzuamb.6
“Nine weeks of trastuzumab was not non-inferior to 1 year of trastuzumab when given with similar chemotherapy,” the researchers concluded. “The docetaxel dosing with trastuzumab requires further study.”6
All of these studies also demonstrated a greater incidence of cardiac toxicity in patients receiving a prolonged dose of trastuzumab, according to Burstein.
“These studies are showing percentages of left ventricle ejection fraction and congestive heart failure that might be of concern,” he told Healio.
In the Short-HER trial, Conte and colleagues concluded that 1 year of trastuzumab should remain the standard treatment in their analysis of 1,254 patients with HER2-positive breast cancer from 82 centers. However, their results showed significantly fewer cardiac events in the 9-week arm compared with the 1-year arm (risk ratio, 0.33; 95% CI, 0.22-0.50).7
Burstein believes such results have “over-magnified” the concern about cardiac events.
“The rejoinder is that a lab-based assessment of these cardiac outcomes is out of touch with what is happening in the clinic,” he said.
Mavroudis and colleagues conducted the HORG trial to examine 6 versus 12 months of trastuzumab therapy, this time in a cohort of 481 women with early, HER2-positive breast cancer. DFS at 3 years served as the primary endpoint.8
The analysis included 240 women in the 6-month group who were followed for 51 months and 241 women in the 12-month group who were followed for 47 months. Disease relapse rates were 11.7% (n = 28) in the 6-month group and 7.1% (n = 17) for the 12-month group (P = .08).8
The results favored 12 months of trastuzumab to 6 months (3-year DFS, 95.7% vs. 93.3%; HR = 1.57; 95% CI, 0.86 to 2.10). OS outcomes were comparable between the two treatment arms, as was cardiac toxicity.8
“Our study failed to show noninferiority for the 6-month arm,” the researchers wrote. “The results further support the current standard of care that is administration of adjuvant trastuzumab for 12 months.”8
All five trials that examined the question of 6 vs. 12 months of trastuzumab had similar results, according to Burstein.
“There is likely not much excitement or interest in asking this clinical question for a sixth time,” he said. “One year of adjuvant trastuzumab therapy should remain the standard of care. None of these trials, in my mind, have successfully swayed that recommendation.” – by Rob Volansky
- Slamon DJ, et al. N Engl J Med. 2011;doi:10.1056/NEJMoa0910383.
- Metzger Filho O & Burstein HJ. Ann Oncol. 2018;doi:10.1093/annonc/mdy480.
- Earl HM, et al. J Clin Oncol. 2018;doi:10.1200/JCO.2018.36.15_suppl.506.
- Hulme C, et al. Ann Oncol. 2018;doi:10.1093/annonc/mdy424.001.
- Pivot X, et al. Cancer Res. 2019;doi:10.1158/1538-7445.SABCS18-GS2-07.
- Joensuu H, et al. JAMA Oncol. 2018;doi:10.1001/jamaoncol.2018.1380.
- Conte P, et al. Ann Oncol. 2018;doi:10.1093/annonc/mdy414.
- Mavroudis D, et al. Ann Oncol. 2015;doi:10.1093/annonc/mdv213.