Revisiting EXTREME in the Immuno‐Oncology Era: How to Improve Its Outcomes

Short abstract

The choice of first‐line palliative treatment has always been one of the crucial decisions in the management of patients with recurrent and/or metastatic squamous cell carcinoma of the head and neck not suitable for salvage surgery or full‐dose radiotherapy. This commentary highlights trial results that show the importance of treatment sequencing decisions that should be of interest for daily practice.

The choice of first‐line palliative treatment has always been one of the crucial decisions in management of patients with recurrent and/or metastatic squamous cell carcinoma of the head and neck not suitable for salvage surgery or full‐dose radiotherapy. Because of the negative impact of cancer progression on overall health status, fewer patients are eligible for systemic approaches later in the disease course, and because of the selection of resistant clones, tumors become less responsive to such therapies [1, 2]. Therefore, we may feel intuitively drawn to opt for the most potent drugs already in the first line. However, growing data underline the importance of treatment sequencing, particularly with respect to immunotherapy and specific drug‐drug interactions including synergism [3]. At present, there are three large phase III trials providing rationale for first‐line decision making, which in turn has consequences for the second line (Table 1) [2, 4, 5].

Table 1.

Phase III trials exploring the EXTREME regimen (PFE) in first‐line treatment of recurrent and/or metastatic squamous cell carcinoma of the head and neck

Trial	Study arms a	Enrollment			Cisplatin, b %			Toxicity, %		Efficacy, mo
		Period	Regions	ITT/SP	Prior	Baseline	Final	G ≥ 3	G5 c	RR, %	PF	OS
EXTREME [4]	PFE	2004–2005	Europe	222/219	41	67	60	83	1	36 d	5.6 d	10.1 d
	PF			220/215	36	61	52	80	3	20	3.3	7.4
KEYNOTE‐048 [2]	PFE	2015–2017	Global	278/287	N/A	44	N/A	83	3	36	5.1	10.7
	KPF			281/276	N/A	43	N/A	85	4	36	4.9	13.0 d , e
TPExtreme [5]	PFE	2014–2017	Europe	270/265	52	100	66	93	4	57	6.2	13.4
	TPE			269/263	58	100	91	81	3	58	6.0	14.5

^a Containing a platinum‐based chemotherapy backbone (single‐agent pembrolizumab arm in KEYNOTE‐048 not mentioned).

^b Prior = patients with previous exposure to platinum; baseline = patients started on cisplatin; final = patients remaining on cisplatin throughout the treatment period (no switch to carboplatin).

^c Treatment‐related deaths.

^d Significant improvement as reported in the respective publication.

^e Benefit driven by patients with tumors harboring cells expressing programmed death ligand‐1.

Abbreviations: E, cetuximab (Erbitux); F, 5‐fluorouracil; G ≥ 3, grade ≥ 3; G5, grade 5; ITT, intention‐to‐treat population (assessment of cisplatin characteristics and efficacy); K, pembrolizumab (Keytruda); N/A, not available; OS, median overall survival; P, platinum; PFS, median progression‐free survival; RR, best objective response to therapy; SP, safety population (assessment of toxicity); T, docetaxel (Taxotere).

In 2008, the results of the EXTREME trial (Erbitux in First‐Line Treatment of Recurrent or Metastatic Head and Neck Cancer) defined a new standard of care consisting of cisplatin at a dose of 100 mg/m² on day 1 plus 5‐fluorouracil at a dose of 1,000 mg/m² on days 1–4 and weekly cetuximab at an initial dose of 400 mg/m², then 250 mg/m². After a maximum of six three‐weekly chemotherapy cycles, patients in the cetuximab group who had at least stable disease received cetuximab monotherapy until disease progression, unmanageable toxicity, or patient refusal [4]. The median number of chemotherapy treatment cycles in the cetuximab group was five, and only 2 of the 100 patients receiving maintenance cetuximab showed an additional response during that period. Despite a breakthrough discovery, this regimen suffered from several shortcomings, including inconvenient continuous administration of 5‐fluorouracil, severe acute adverse events occurring in the majority of patients, and a lack of predictive biomarkers. Moreover, fewer than 5% of patients were still alive at 5 years [6]. Nevertheless, it represented the first validated regimen based on level 1 evidence, and its efficacy was afterward repeatedly reconfirmed in retrospective studies [7].

It took more than 10 years to change the paradigm. The three‐arm KEYNOTE‐048 trial showed better outcomes with introduction of immunotherapy to the first line supported by identification of a new predictive biomarker [2]. The expression of programmed death ligand‐1 (PD‐L1) calculated as combined positive score (CPS) was associated with improved overall survival in patients receiving pembrolizumab, either alone or in combination with a chemotherapy backbone, whereas patients with a low expression (CPS <1) showed no benefit (in case of the combination with chemotherapy) or worse outcome (in case of monotherapy) [2, 8]. Concerns were raised also in patients presenting with locoregional recurrence only. In the two experimental arms with the immune checkpoint inhibitor, the rates of progression were consistently higher compared with that observed with EXTREME, likely leading to the lack of survival benefit observed in cases with an expansive process only in the craniocervical region, known for the proximity of vital structures [9]. This hypothesis is further supported by the existence of a perilous phenomenon characterized by an accelerated rate of tumor growth stimulated by immunotherapy. This so‐called hyperprogression is now accepted by most oncologists despite initial doubts about possible confounding with natural history of the disease [10]. Therefore, EXTREME followed by immunotherapy in the second line might be a viable option in these cases. Another supporting argument relates to the likely negative correlation between cancer burden and response to immunotherapy because large tumors seem to respond less well than small tumors [11]. For that reason, initial chemotherapy may serve to induce size reduction, being especially appropriate if anatomical space tolerance is limited.

Most recently, EXTREME was challenged by TPExtreme. Based on preclinical research and single‐arm prospective trials, emerging evidence suggests a synergistic effect of cetuximab‐taxane combinations [3]. In the experimental arm (TPEx) of the phase III TPExtreme study (GORTEC 2014‐01), 5‐fluorouracil was replaced by docetaxel. Besides that, only four chemotherapy cycles were planned, the dose of cisplatin was reduced from 100 mg/m² to 75 mg/m², and in the maintenance phase, the dose of cetuximab was doubled and given every 2 weeks [5]. It had already been demonstrated that 5‐fluorouracil can be replaced by a taxane in combination with cisplatin with the same efficacy but lower toxicity [12]. Therefore, TPExtreme sought to find if the alleged synergism between cetuximab and taxanes could be translated into better survival and tumor shrinkage in comparison with EXTREME. Although these endpoints were not met, the trial brought important new data on the use of EXTREME in the first‐line setting as well as on TPEx as its alternative. Notwithstanding the negative statistical outcome of this superiority trial, the convincing numerical results that would have met criteria for noninferiority if such a design had been chosen, provide sufficient arguments to endorse TPEx in situations where contraindications to 5‐fluorouracil and/or its continuous administration exist if adequate supportive care can be assured. Although TPEx had an overall lower toxicity with, for example, a decreased frequency of acute mucositis (46% vs. 58%; grade 3 or more: 8% vs. 13.5%), it led to more febrile neutropenia (10% vs. 6%; treatment‐related deaths: three patients vs. none) despite an obligatory granulocyte colony‐stimulating factor (G‐CSF) support. This finding will probably have implications for geographical regions with limited access to intensive care that is due either to a shortage in its availability, such as during the COVID‐19 pandemic, or to longer travel distances.

The first observation of the TPExtreme study that can be translated into clinical practice to optimize the traditional EXTREME regimen is the dosing schedule of cetuximab maintenance, which can be prescribed biweekly instead of weekly. This modification has recently been approved by the Food and Drug Administration but had already been adopted by many institutions prior to the TPExtreme publication [13]. Next, an exploratory subgroup analysis of patients treated with EXTREME showed that the use of G‐CSF was associated with a survival benefit. Although growth factor support in the palliative setting remains a controversial topic, as no prospective randomized phase III trials have shown survival benefit with G‐CSF in this setting, it might represent a feasible strategy to reinforce the outcomes in fit patients in whom we aim for maximum downsizing because of disease symptoms or the possibility of subsequent local consolidation [3]. Finally, the TPExtreme trial provides valuable new data on the role of platinum compounds in the EXTREME regimen. In general, carboplatin represents a better tolerated alternative to cisplatin, albeit less efficacious, in terms of both survival and response rates [3]. The 2008 EXTREME trial and the KEYNOTE‐048 study allowed investigators to make a choice between the two drugs. However, this was not the case of the TPExtreme trial, where all participants must have started on cisplatin (Table 1). Such increased exposure to cisplatin could have contributed to superior outcomes of the EXTREME regimen in the TPExtreme trial relative to former trials, demonstrating thus how this treatment schedule can be further improved.

Owing to cumulative toxicity, limitations have been imposed on the maximum lifetime dose of cisplatin. Previous administration of more than 200 mg/m² can be considered a relative contraindication for retreatment with cisplatin [14]. The same cutoff has been used to define the minimum sufficient dose of concurrent cisplatin during chemoradiotherapy of locoregionally advanced head and neck cancer since higher exposure to cisplatin increases the risk of toxicity with an equivocal impact on overall survival [15]. In the TPExtreme study, about half of patients had been pretreated with a platinum agent, and the planned total doses of cisplatin were 600 mg/m² in the EXTREME arm but only 300 mg/m² in the TPEx arm. This imbalance was partially responsible for different toxicity profiles between the two study groups as reflected by more frequent dose adaptations and replacements of cisplatin by carboplatin in the EXTREME arm. Moreover, only about half of patients randomized to EXTREME in the TPExtreme study could receive more than four chemotherapy cycles. Hence, we may speculate that the toxicity differences could have repercussions on the nonsignificant separation of the overall survival curves disadvantaging EXTREME.

Consequently, the results of the TPExtreme trial may help us revise the standard EXTREME regimen, which remains pivotal for patients who cannot benefit from immune checkpoint inhibitors because of low PD‐L1 expression or limited access to immunotherapy in resource‐restricted regions. The optimization of different aspects of outcome, including efficacy, toxicity, and compliance, could be accomplished by decreasing the frequency of cetuximab administrations in the maintenance phase, implementing G‐CSF in selected patients to retain the maximum dose intensity, and reducing the number of chemotherapy cycles from six to four, particularly with respect to cisplatin, or altering the administration schedule of cisplatin and lowering the peak level in the blood, which correlates with its acute toxicities [16, 17]. Studies evaluating such down‐to‐earth adaptations of the EXTREME regimen seem of interest for daily practice.

Disclosures

Petr Szturz: Merck‐Serono, Servier, Bristol‐Myers Squibb (C/A); Jan B. Vermorken: Innate Pharma, Merck Sharp & Dohme, Corp., PCI Biotech, Nanobiotix, WntResearch (C/A), Merck‐Serono, Merck Sharp & Dohme, Corp., Bristol‐Myers Squibb (H).

(C/A) Consulting/advisory relationship; (RF) Research funding; (E) Employment; (ET) Expert testimony; (H) Honoraria received; (OI) Ownership interests; (IP) Intellectual property rights/inventor/patent holder; (SAB) Scientific advisory board