Spotlight on landmark oncology trials: the latest evidence and novel trial designs.

The era of precision oncology is marked with prominent successes in the therapy of advanced soft tissue sarcomas, breast cancer, ovarian cancer and haematological neoplasms, among others. Moreover, recent trials of immune checkpoint inhibitors in melanoma, non-small cell lung carcinoma, and head and neck cancers have significantly influenced the therapeutic landscape by providing promising evidence for immunotherapy efficacy in the adjuvant setting in high-risk locoregional disease. To speed up the introduction of targeted therapy for cancer patients, novel phase II trials are being designed, and may likely form the basis for the 'landmark trials' of the future. A special article collection in BMC Medicine, "Spotlight on landmark oncology trials", features articles from invited experts on recent clinical practice-changing trials.

Background

The era of precision medicine has led to significant developments in the therapy of advanced soft tissue sarcomas (STS), breast cancer, ovarian cancer and haematological neoplasms, among others. However, cancer research also faces challenges in the effective development and assessment of targeted therapeutics [1], including the need for early evaluation of potential biomarkers by translational and correlative studies.

In this editorial, we discuss the special article collection entitled “Spotlight on landmark oncology trials” recently published in BMC Medicine, which focuses on the core clinical trials of selected solid tumours (lung cancer [2], melanoma [3, 4], STS [5], head and neck cancer [6]). We also highlight selected and recent practice-changing trials in chronic lymphocytic leukaemia as well as breast and gynaecological cancers, and review the advances offered by the development of novel clinical trial designs.

Recent landmark immunotherapy trials – melanoma, non-small cell lung carcinoma (NSCLC), head and neck cancer

The first articles in the special article collection focus on landmark clinical trials in selected advanced solid tumours, with special attention on the most studied tumours with regards to immunotherapy development, namely melanoma [3, 4], NSCLC [2], and head and neck cancer [6]. Recent developments and approvals in immunotherapy have significantly changed the landscape of melanoma and NSCLC therapy in the metastatic setting, and open various possibilities for adjuvant treatment in high-risk locoregional disease [7-10]. In this article series, worldwide renowned experts in their fields provided an extensive overview on the state of the art in immunotherapy and discussed the possible future paths in these, still difficult, types of malignancies.

The current results of anti-PD-1 therapy with pembrolizumab or nivolumab monotherapy in melanoma indicated a median overall survival (OS) of approximately 2 years, but the combination of anti-PD-1 and anti-CTLA-4 (nivolumab with ipilimumab) was shown to be superior in terms of progression-free survival (PFS) and OS (Table 1) [11-15]. Further clinical trials are under way to determine how best to integrate combination immunotherapy and other treatment modalities as well as to establish the correct sequence of therapy with targeted treatment in BRAF-mutated cases.

Table 1

Summary of major trials with immunotherapy in advanced cutaneous melanoma

Study	N (% 1st line)	ORR	Median PFS	2-year OS rate
Pembrolizumab (10 mg/kg every 2 weeks)KEYNOTE-006Robert et al., 2015 [11]Schachter et al., 2016 [12]	279 (65.6%)	33.7%	5.6 months	55.0%
Pembrolizumab (10 mg/kg every 3 weeks)KEYNOTE-006 Robert et al., 2015 [11]Schachter et al., 2016 [12]	277 (66.8%)	32.9%	4.1 months	55.0%
Nivolumab in BRAF-negative melanoma Checkmate-066 Robert et al., 2015 [13]	210 (100%)	40.0%	5.4 months	57.7%
Ipilimumab + Nivolumab Checkmate-067 Larkin et al., 2015 [14]Larkin et al., 2017 [15]	314 (100%)	58.9%	11.7 months	64.0%

ORR overall response rate; OS overall survival; PFS progression-free survival

The team lead by Professor Jean-Charles Soria discussed the successes and failures of immunotherapy in the first-line treatment of NSCLC [2]. Moreover, three anti-PD-1/anti-PD-L1 agents, pembrolizumab, nivolumab and atezolizumab, have been approved for second-line therapy of NSCLC [16-18]; however, contrary to melanoma, patient selection to therapy should be based on PD-L1 expression level of tumour cells.

Recent landmark trials in STS

Another topic featured in this article collection is systemic therapy in STS [5], which is a heterogeneous group of rare solid tumours. Despite optimal local treatment, approximately 50% of adult patients with localised STS develop distant metastases and die of metastatic disease. A limited number of drugs have shown activity in advanced disease, and due to the rarity of these tumours, clinical trials in sarcoma include many subtypes and are mainly initiated by academic research groups.

Recent developments in the classification of STS, insights into their molecular pathogenesis and the optimal treatment strategies have evolved considerably during the past decades and have led to the introduction of new therapies. Nevertheless, the selection of systemic therapy must be strictly individualised and based upon several factors, including the histology and biological behaviour of the disease. A summary of recent pivotal trials for systemic therapy in advanced STS is presented in Table 2 [19-22].

Table 2

Summary of recent pivotal clinical trials in advanced soft tissue sarcomas

Tumour type, phase [reference]	Line of therapy	Arms (experimental vs. control)	Response rate	Clinical benefit rate	Median PFS (months, P value)	Median OS (months, P value)
Non-adipocytic soft tissue sarcoma, phase 3, n = 369 [19]	Second or later (after anthracycline)	Pazopanib 800 mg/m2 vs. placebo	6% vs. 0%	73% vs. 38%	4.6 vs. 1.6 (P < 0.0001)	12.5 vs. 10.7 (P = 0.25)
Liposarcoma and leiomyosarcoma, phase 3, n = 518 [20]	Second or later (after anthracycline)	Trabectedine 1.5 mg/m2 vs. dacarbazine 1000 mg/m2	10% vs. 7%	61% vs. 42%	4.2 vs. 1.5 (P < 0.001)	12.4 vs. 12.9 (P = 0.37)
Liposarcoma and leiomyosarcoma, phase 3, n = 452 [21]	Third or later (after anthracycline)	Eribulin mesylate 1.4 mg/m2 vs. dacarbazine 850–1200 mg/m2	5% vs. 4%	57% vs. 52%	2.6 vs. 2.6 (P = 0.23)	13.5 vs. 11.5 (P = 0.01)
Soft tissue sarcoma, phase 2, n = 133 [22]	First line	Olaratumab 15 mg/kg plus doxorubicin 75 mg/m2 vs. doxorubicin alone 75 mg/m2	18.2% vs. 11.9%	77.3% vs. 62.7%	6.6 vs. 4.1 (P = 0.06)	26.5 vs. 14.7 (P = 0.0003)

OS overall survival; PFS progression-free survival

Recent landmark trials in breast and gynaecological cancers

Recent landmark trials in HER2-positive breast cancer include those using dual HER2-targeted therapy pertuzumab and trastuzumab with docetaxel. In the neoadjuvant setting, the NeoSphere trial demonstrated significantly improved pathological complete response rates [23] and a trend favouring improved PFS and OS at 5 years [24]. Results from the CLEOPATRA trial in the metastatic setting of the same treatment have produced remarkable results [25]; the same combination produced a 56.5-month median OS compared with 40.8 months achieved with trastuzumab and docetaxel alone, showing an increase of 15.7 months to OS in the pertuzumab group. These results clearly demonstrate the superiority of dual HER2-directed therapy. In ER-positive, HER2-negative metastatic disease, the landmark trial (PALOMA 3) uses the CDK 4/6 inhibitor, palbociclib [26, 27]. Median PFS was 9.5 months in the fulvestrant plus palbociclib group and 4.6 months in the fulvestrant plus placebo group with a hazard ratio of 0.46, which was highly statistically significant. However, translational research did not discover any predictive biomarker subgroups [27] for the palbociclib effect.

The landmark phase III trials in high-grade serous ovarian cancer are testing PARP inhibitors as maintenance therapy after response to platinum-based therapy in relapsed disease. Study 19 [28, 29] used olaparib against placebo and demonstrated a PFS of 11.2 months in BRCA-mutated patients compared with 4.3 months for wild-type patients (hazard ratio, 0.18; P < 0.0001). A more recent niraparib study had similar results [30], where patients in the niraparib group had a significantly longer PFS than the placebo group in all cohorts tested (21.0 vs. 5.5 months in the gBRCA cohort; 12.9 vs. 3.8 months in the non-gBRCA cohort for patients who had tumours with homologous recombination deficiency; and 9.3 vs. 3.9 months in the overall non-gBRCA cohort; P < 0.001). Both trials demonstrated significant benefit for maintenance PARP inhibitors in all subgroups of platinum-sensitive relapsed high-grade serous ovarian cancer.

Recent landmark trial in chronic lymphocytic leukaemia (CLL): upfront therapy with ibrutinib in elderly with chronic lymphocytic leukaemia (The RESONATE-2 Trial)

With the advent of novel oral agents that are well tolerated and highly efficacious, the therapeutic landscape of CLL underwent radical changes [31]. In phase 3 trials, ibrutinib, a first-in-class Bruton tyrosine kinase (BTK) inhibitor, showed efficacy over traditional salvage therapeutic options in patients with relapsed or refractory CLL [32]. The importance of BTK inhibitors in the first-line setting has been recently investigated in the RESONATE-2 study [33], a head-to-head clinical trial in which outcomes were shown to be superior for patients who received ibrutinib in comparison to patients treated with chlorambucil single agent. Indeed, ibrutinib demonstrated a survival advantage over chlorambucil despite the study’s crossover design. The strength of the study also relies on the good tolerance profile of ibrutinib, which allows it to be administered continuously and provide indefinite disease suppression even in elderly or unfit CLL patients. However, a potential setback is represented by the control arm since chlorambucil is no longer regarded an adequate therapy in CLL [26]. Combinations of chlorambucil with an anti-CD2O monoclonal antibody, such as rituximab, ofatumumab or obinutuzumab, are now the standard of care in patients unsuited to receive fludarabine, cyclophosphamide and rituximab [34-36]. Further, a trial comparing ibrutinib plus obinutuzumab versus chlorambucil plus obinutuzumab finished recruiting, but results are not yet mature (ClinicalTrials.gov Identifier: NCT02264574). Therefore, in absence of data from this and similar trials, either therapeutic choice is adequate in the day-to-day practice.

In conclusion, the RESONATE-2 trial demonstrates that ibrutinib is a new important player in the treatment of elderly unfit patients and in those with high-risk disease. A potential shortcoming with the upfront use of ibrutinib includes cost and indefinite treatment course [37]. Finally, biomarker and minimal residual disease assessment may ultimately be useful to guide the targeted agent or regimen of choice and the duration of treatment [38].

Novel trial designs and conclusions

The landmark oncology trials highlighted in the BMC Medicine series “Spotlight on landmark oncology trials” and this editorial are recent trials that have produced practice-changing results for patients. These trials represent the end of the long process of translating scientific innovation and drug discovery, through first-in-man studies, followed by phase II trials and finally by randomised phase III trials as required for licensing of new treatments. Novel trial designs could potentially lead to a different type of landmark trial that would accelerate the process and allow cancer patients to access new treatments faster. In the phase I setting, there is a pressing need to develop better trial methodologies for novel combinations, often of a standard chemotherapy with a novel targeted agent. Di Veroli et al. [39] published an interesting software to provide information in terms of synergy and/or antagonism between two compounds. In addition, adaptive designs for phase I combinations are being developed [40].

In the era of precision cancer medicine, innovative trial designs will also require the matching of novel drugs with putative targets. Indeed, BATTLE, a landmark phase II trial using an adaptive randomised design, tested four novel drugs and biomarker pairings in NSCLC [41]. This was followed by BATTLE-2 [42], testing combination treatments in the same disease. In neoadjuvant breast cancer, the I-SPY 1 and 2 trials have successfully matched treatment and biomarkers, using adaptive randomised designs [43, 44]. Landmark results include those in triple negative breast cancer for the combination of velaparib and carboplatin [44] and neratinib in HER2-positive breast cancer [45]. The FOCUS 4 trial in metastatic colorectal cancer uses group-sequential multi-arm, multi-stage methodology [46] to achieve similar matching of novel therapy and biomarker groups. Our own group is developing a novel Bayesian, adaptive randomised methodology [47]. These designs would allow recruitment of biomarker-negative patients, often not included in other trials, and have the potential for ‘perpetual’ designs, in which successful matching of novel drugs and biomarkers would result in ‘graduation’ of the pair to a phase III trial, along with the rapid rejection of novel drugs that did not work. More effective and cost-efficient phase II trial designs would rapidly lead to landmark trials and practice-changing results.