HER2阳性乳腺癌：紫杉类选哪种？
1998年，曲妥珠单抗被批准联合化疗（当时为蒽环类和环磷酰胺或紫杉醇）用于HER2阳性晚期乳腺癌，为靶向药物广泛用于抗癌治疗空前发展的新时代铺平了道路。2009年，随着曲妥珠单抗与帕妥珠单抗协同作用的临床前研究证据发表，双重HER2靶向治疗进一步改变了HER2阳性晚期乳腺癌患者的治疗方法。2012年，改变临床实践的CLEOPATRA研究结果表明，曲妥珠单抗＋多西他赛＋帕妥珠单抗与曲妥珠单抗＋多西他赛相比，中位总生存可延长16个月。该研究重新定义了我们的日常实践，该联合方案被迅速用于HER2阳性晚期乳腺癌患者的一线治疗。虽然每周紫杉醇与每3周多西他赛相比，有效性结局相似，但是急性毒性（中性粒细胞减少和中性粒细胞减少伴发热）发生率显著较低，故该方案被许多国家首选。尤其对于体弱和老年患者，临床医师倾向采用每周紫杉醇。此外，新开发的白蛋白紫杉醇不须溶剂，可避免紫杉醇和多西他赛所须合成溶剂以及类固醇预用药相关严重毒性，目前已被许多国家批准，但是与传统紫杉类相比，对于晚期乳腺癌的疗效并不更好。考虑到这些因素，CLEOPATRA研究结果的发表，提出了紫杉醇或白蛋白紫杉醇可否有效取代多西他赛的问题。
CLEOPATRA研究结果发表后，欧洲启动了PERUSE研究，对曲妥珠单抗＋帕妥珠单抗＋不同紫杉类（多西他赛、紫杉醇、白蛋白紫杉醇）用于HER2阳性晚期乳腺癌日常临床实践的安全性和有效性进行了比较，并于2019年2月23日和2021年7月1日先后公布了初步结果和最终结果。PERUSE研究共入组1436例患者，是评估曲妥珠单抗＋帕妥珠单抗＋不同紫杉类的最大研究。紫杉类由研究者自行选择，被选择最多的紫杉类为多西他赛（775例，54%），其次为紫杉醇（588例，41%，大多每周给药）、白蛋白紫杉醇（65例，5%）。结果，紫杉类的选择似乎并未影响生存结局。虽然PERUSE研究设计目的并非正式比较不同紫杉类，但是多西他赛、紫杉醇、白蛋白紫杉醇相比，患者的中位无进展生存（19.4、23.2、19.2个月）和总生存（66.5、64.0、70.9个月）都相似，说明紫杉醇或白蛋白紫杉醇都以有效取代多西他赛。正如预期，老年患者和症状负担较高的患者接受紫杉醇治疗比例较高。不过，紫杉醇与多西他赛相比，患者发生≥3级中性粒细胞减少和中性粒细胞减少伴发热风险显著较低，尽管作者未报告紫杉类总剂量和粒细胞巨噬细胞集落刺激因子用药情况。有趣的是，无论何种紫杉类，患者年龄≥65岁与<65岁相比，紫杉类中位疗程都为大约4个月。这些数据似乎不同于CLEOPATRA研究报告的患者年龄≥65岁与<65岁相比多西他赛中位疗程较低（6比8轮）。因此，我们能够质疑根据患者特征选择最合适的紫杉类或改用其他紫杉类（占4%的入组患者）可否避免老年患者次优治疗和早期治疗中断。
PERUSE研究的意义不仅限于选择最佳紫杉类联合双重HER2靶向治疗，还应赞扬作者通过严格和务实的方法挑战了CLEOPATRA研究结果对于临床实践的可转化性。PERUSE研究与CLEOPATRA研究相比，入组患者更好地反映了我们日常临床实践接受治疗的患者。PERUSE研究与CLEOPATRA研究相比，曲妥珠单抗术前新辅助或术后辅助治疗患者入组比例较高（29%比12%），激素受体阳性晚期乳腺癌较多（64%比49%）并被允许接受维持内分泌治疗（21%）。由于既往接受过曲妥珠单抗（新）辅助治疗的患者较多，作者能够证实与未接受过曲妥珠单抗治疗的患者相比，中位无进展生存（15.4比23.4个月）和中位总生存（54.1比73.5个月）显著较短，这些结果与一些临床实践证据一致。该不同影响可能由于曲妥珠单抗耐药以及初诊四期乳腺癌与复发乳腺癌之间固有临床和生物学差异所致。PERUSE研究与CLEOPATRA研究相比，虽然存在不良结局风险的患者比例较高，但是中位总生存较长（65.3比57.1个月），尽管这是根据间接比较。该生存结局改善主要由于有效HER2靶向药物不断增加及其对临床实践报告生存结局的有利影响。此外，PERUSE研究与CLEOPATRA研究相比，由于接受帕妥珠单抗＋曲妥珠单抗治疗的患者人数增加3.5倍，有可能检测到少见的长期治疗相关毒性。令人欣慰的是，最终分析与初步分析相比，收集的安全数据相似，并且未发现累积毒性或新的安全问题出现。
不过，许多问题仍未得到解决。首先，PERUSE研究正式允许入组中枢神经系统转移患者，相比之下，此类患者被CLEOPATRA研究排除。然而，考虑到入组仅限于局部治疗后稳定≥3个月且未经任何HER2靶向治疗的中枢神经系统转移患者，此类患者数量可能极少，故无法得出任何关于该特定情况治疗效果的明确结论。其次，虽然整个治疗期间通过癌症治疗功能评定乳腺癌问卷（FACT-B）广泛收集患者报告结局，但是并未根据不同紫杉类对数据单独分析。事实上，随着HER2阳性晚期乳腺癌患者生存率显著提高，考虑不同紫杉类对生活质量的长期影响变得越来越重要。最后，由于目前曲妥珠单抗＋帕妥珠单抗、恩美曲妥珠单抗现被广泛用于治疗早期乳腺癌，并且PERUSE研究开展时间远早于APHINITY和KATHERINE研究结果公布，故这些结果对此类患者的普遍适用性仍然有待解答。
相关链接
Ann Oncol. 2021 Aug 13. Online ahead of print.
Are all taxanes equal in patients with HER2-positive breast cancer? PERUSing the safety and efficacy of trastuzumab plus pertuzumab and taxanes in real-life usual care.
Rassy E, Pistilli B.
Gustave Roussy, Villejuif, France.
"In Faith, I will. Let me peruse this face"
- William Shakespeare, Romeo and Juliet-Act V
Almost a quarter century ago, the addition of trastuzumab to a chemotherapy regimen (anthracyclines and cyclophosphamide or paclitaxel at the time) paved the way for a new era in the treatment of human epidermal growth factor receptor 2 (HER2)-positive breast cancer and more widely for an unprecedented development of anti-cancer targeted therapies. A decade later, the preclinical evidence of a synergy between trastuzumab and pertuzumab led to a dual HER2-targeting approach that transformed the treatment of patients with HER2-positive advanced breast cancer (ABC). Indeed, the practice-changing CLEOPATRA trial showed that the addition of pertuzumab to trastuzumab and docetaxel improved survival (OS) by 16 months in comparison to trastuzumab plus docetaxel. This trial redefined our daily practice and this combination was rapidly adopted in the first-line treatment of patients with HER2-positive ABC. While efficacy outcomes are similar between 3-weekly docetaxel and weekly paclitaxel, the incidence of acute toxicities, namely neutropenia and neutropenic fever, is remarkably lower with the latter, thus making this regimen the preferred choice in many countries. Particularly, clinicians tend to favor weekly paclitaxel among frail and elderly patients. Furthermore, nab-paclitaxel, a solvent-free formulation of paclitaxel, was later developed to avoid the severe toxicities associated with the synthetic solvents contained in paclitaxel and docetaxel and the premedication with steroids. Its use is currently authorized in many countries, although it did not show superior efficacy outcomes in comparison to conventional taxanes in the treatment of ABC. Given these considerations, the publication of the CLEOPATRA results raised the question whether docetaxel can be effectively replaced by paclitaxel or nab-paclitaxel.
In this issue of Annals of Oncology, Miles and colleagues report the final results of the phase IIIb PERUSE (PERtUzumab global SafEty) study which was designed to assess the safety and efficacy of pertuzumab plus trastuzumab and an investigator-selected taxane (docetaxel, paclitaxel, and nab-paclitaxel) in patients with HER2-positive ABC. With 1,436 patients, PERUSE is the largest study that evaluated the combination of trastuzumab plus pertuzumab and any of the 3 taxanes in this setting. Based on the investigators' choice, docetaxel was the most commonly selected taxane (n = 775; 54%), followed by paclitaxel (n = 588; 41%), administered weekly in most cases, and nab-paclitaxel (n = 65; 5%). Crucially, the choice of the taxane agent did not seem to affect the survival outcomes. Although the PERUSE study was not designed to formally compare the 3 subgroups, progression-free survival (PFS; 19.4, 23.2, 19.2 months, respectively) and OS (66.5, 64.0, 70.9 months, respectively) were similar among patients who received docetaxel, paclitaxel or nab-paclitaxel, indicating that docetaxel can be effectively replaced by paclitaxel or nab-paclitaxel. As expected, older patients and those with a higher symptomatology burden had a higher likelihood of receiving paclitaxel. Notwithstanding, patients treated with paclitaxel had a substantially lower risk of grade ≥ 3 neutropenia and febrile neutropenia in comparison to those who received docetaxel, although the authors did not report the cumulative dose of taxanes and the use of G-CSF. Interestingly, there was a striking similarity in median taxane exposure between patients older than 65 and their younger counterparts (≈ 4 months), regardless of the taxane agent. These data seem to differ from those reported in the CLEOPATRA trial, where the median number of docetaxel cycles was lower in patients older than 65 years (6 cycles vs 8 cycles). Therefore, we can argue that the selection of the most appropriate taxane upon patient's characteristics or switching to an alternative taxane (reported in 4% of the enrolled patients) may avoid suboptimal therapy and early treatment discontinuation in elderly patients.
The relevance of the PERUSE study is not limited to the choice of the optimal taxane agent in combination with the dual HER2-targeted therapy. The authors should also be praised for having challenged the transferability of the CLEOPATRA results in the real-life setting by using a rigorous and pragmatic approach. In comparison to the CLEOPATRA trial, the patients enrolled in the PERUSE study better mirror those treated in our daily clinical practice. The PERUSE study enrolled a higher proportion of patients that had received (neo)adjuvant treatment with trastuzumab (45% of patients initially diagnosed with early BC and 29% of the overall population compared with 12% in CLEOPATRA), most patients had HR-positive ABC (64% vs 49%) and were allowed to receive maintenance endocrine therapy (21%). With a larger inclusion of patients previously treated with adjuvant trastuzumab, the authors could demonstrate a shorter PFS and OS in patients previously exposed to HER2-targeted therapy in comparison to their trastuzumab-naive counterparts (15.4 vs 23.4 months and 54.1 vs 73.5 months, respectively). These findings are consistent with some real-life evidence that reported worse efficacy outcomes in patients receiving trastuzumab, pertuzumab and taxane after previous (neo)adjuvant trastuzumab. This differential impact can be driven by the onset of trastuzumab-induced resistances as well as by the intrinsic clinical and biological differences between de novo ABC and recurrent breast cancer. Despite the higher prevalence of patients at risk of poorer outcomes, overall survival of PERUSE trial exceeded that of CLEOPATRA trial (65.3 months as compared to 57.1 months), though by indirect comparison. This improvement mostly derives from the growing armamentarium of effective HER2-targeted therapies and their favorable impact on survival outcomes as reported in real-life settings. Furthermore, with the enrollment of 3.5 fold the number of patients in the CLEOPATRA trial who received pertuzumab plus trastuzumab, PERUSE had the potential of detecting rare and long-term treatment-related toxicities. So it is reassuring that the safety data collected in the final analysis paralleled those of the preliminary analysis and did not reveal cumulative toxicities or the emergence of new safety signals.
A number of questions remain unaddressed. First, the PERUSE study formally allowed the inclusion of patients with CNS metastases that, by contrast, were excluded from the CLEOPATRA trial. However, considering that the enrollment was restricted only to patients with CNS metastases that were stable for ≥ 3 months after local therapy and without any HER2-targeted therapy, it is likely that the number of these patients was remarkably low, thus precluding any definite conclusions regarding the treatment efficacy in this specific setting. Second, while patient-reported outcomes were extensively collected through the FACT-B questionnaires over the whole treatment duration, data were not separately analyzed according to the type of taxane. Indeed, with the major improvements in survival in patients with HER2-positive ABC, it is becoming increasingly important to consider the long-term impact of different taxanes on quality of life. Finally, as trastuzumab-pertuzumab combinations and T-DM1 are now more widely used for the treatment of early breast cancer and PERUSE trial was conducted far earlier than the publication of APHINITY and KATHERINE results, the generalizability of these results to these patients remains an open question.
DOI: 10.1016/j.annonc.2021.08.1750