The primary challenge of mRCC is that complete response to treatment with a single agent is rare. Disease progression is expected and tumor resistance is an inevitable reality. mRCC remains incurable in most instances, and mechanisms of tumor-cell resistance to conventional radiotherapy and chemotherapy have been an active area of research. Proposed mechanisms include overexpression of multidrug resistance gene
MDR-1, cell survival gene
clusterin, PKC-ζ, L2 cell adhesion molecule L1-CAM, P-glycoprotein, various DNA repair proteins, the antiapoptotic gene
bcl-2, glutathione S-transferase, decreased expression of DNA topoisomerase, loss of HIF-1α regulation, accumulation of HIF-2α, and suppression of p53 [
40,
42]. Targeting these resistance mechanisms is an area of ongoing studies and may play a role in future approaches to the treatment of advanced RCC.
Prior to the approval of everolimus by FDA in 2009 for the second-line use in mRCC, there was no established treatment option for patients who progressed on first-line VEGF-directed therapy. Significant advancement has been made since, and the second-line treatment options now include TKIs—axitinib, cabozantinib, and lenvatinib; an anti-PD1 monoclonal antibody, nivolumab; and an mTOR inhibitor, everolimus. Sequential treatments have emerged as a viable approach to controlling drug resistance and overcoming resistance mechanisms [
43], while the optimal sequence of treatment remains to be defined as few studies to date have directly compared drug efficacy [
44].
Everolimus
The oral agent mTOR inhibitor, everolimus, was the first drug to be approved for second-line use in mRCC after progression on first-line VEGF TKI treatment. In the RECORD-1 trial, everolimus was compared to placebo in 410 mRCC patients whom had been previously treated for at least 6 months with sunitinib, sorafenib, or both [
45]. The study’s primary endpoint was mPFS, and everolimus performed superiorly to placebo with a significant mPFS difference of 4.0 to 1.9 months, respectively. A follow-up trial, RECORD-4, prospectively followed mRCC patients on everolimus after the progression of disease on either sunitinib, additional VEGF TKIs, or cytokine therapy [
46] and confirmed the mPFS benefit of everolimus. The role of everolimus has been investigated beyond the second-line setting and investigators proposed that mTOR inhibition may have a role in untreated nccRCC. The ASPEN trial was a multicenter, open-label, randomized phase II trial designed to determine the mPFS of sunitinib and everolimus in a subset of patients with histologically proven nccRCC. Results favored sunitinib over everolimus (8.3 vs. 5.6 months) and reaffirmed the role of everolimus solely as a second-line agent, irrespective of histological subtype [
47].
Sorafenib and axitinib
The role of second-line sorafenib has been investigated in the INTORSECT trial, which enrolled patients whom had progressed after sunitinib therapy and randomized them to receive either temsirolimus or sorafenib with a primary endpoint of mPFS and secondary endpoints of safety, ORR and OS. The study concluded there was no difference in mPFS between the two agents; however, sorafenib demonstrated a superior OS of 16.6 months over temsirolimus, 12.3 months. Both agents presented with an acceptable safety profile, and the adverse effects were consistent with the known toxicities of the drugs [
48].
Axitinib is an oral, potent, small-molecule TKI that selectively inhibits VEGFR-1, VEGFR-2, and VEGFR-3 [
49]. The role of axitinib in the second-line treatment of mRCC has been investigated in the phase III AXIS trial, which randomized 723 patients with mRCC whom had disease progression after first-line systemic therapy [
41]. First-line therapy included sunitinib, cytokine therapy, bevacizumab plus IFN-α, or temsirolimus. The trial met its primary endpoint, mPFS, and axitinib was shown to offer a superior mPFS of 6.7 months compared to sorafenib, 4.7 months. There was, however, no difference in the OS between the axitinib-treated group and sorafenib-treated group [
50]. Common axitinib toxicities observed in the study included diarrhea (55%), hypertension (40%), and fatigue (39%).
Cabozantinib
The second-line agents everolimus and axitinib had become the standard of care in refractory disease, but the mPFS was only extended by a mere 3 to 5 months after disease advancement on first-line therapy [
41,
48]. Within the past year, two novel VEGFR TKIs, cabozantinib and lenvatinib, have gained FDA approval for use in advanced RCC.
Cabozantinib is an oral, small-molecule TKI-targeting VEGFR that was originally approved for metastatic medullary thyroid cancer. In addition to VEGFR, cabozantinib targets receptor tyrosine kinases implicated in and relevant to mRCC; RET, KIT, AXL, and FLT3 [
51]. A landmark study by Zhou et al. provided evidence that MET and AXL are upregulated in chronic sunitinib use and play a role in RCC tumor resistance to TKIs [
52]. This data is in concordance with previous studies which suggest poor prognosis when MET/AXL are highly expressed by RCC tumor cells [
53]. Cabozantinib was studied in the METEOR trial, which was a randomized, open-label, phase III trial comparing cabozantinib with everolimus in 658 patients with mRCC whom had advanced after TKI therapy. The study’s primary endpoint was mPFS and secondary endpoints were OS and ORR. The rate of disease progression with cabozantinib was 42% lower than with everolimus. The METEOR trial met its primary endpoint, and cabozantinib demonstrated a superior mPFS of 7.4 months compared to 3.8 months with everolimus. An OS advantage was observed with cabozantinib (21.4 compared to 16.5 months), and the ORR significantly favored cabozantinib to everolimus, 21 to 5% (
p < 0.001) [
38]. Cabozantinib was observed to have a similar safety profile to drugs in its own class (Table
1). The incidence of grade 3 and 4 adverse effects was 68% with cabozantinib. The most common events were hypertension (15%), diarrhea (11%), and fatigue (9%). Dose reductions occurred in 60% of the study patients stratified to cabozantinib treatment. A grade 5 adverse event occurred in one patient [
54].
Table 1
Common adverse effects of novel agents approved for mRCC
Diarrhea | 85 | 11 | 13 | 1 | 72 | 12 |
Fatigue | 65 | 9 | 35 | 2 | 50 | 8 |
Arthralgia/myalgia | 11 | <1 | (11–21) | (0) | 25 | 0 |
Decreased appetite | 48 | 2 | 12 | <1 | 58 | 4 |
Vomiting | 34 | 2 | (15–17) | (0) | 39 | 4 |
Nausea | 54 | 4 | 14 | <1 | 62 | 8 |
Stomatitis | 24 | 2 | 2 | 0 | 25 | 2 |
Hypertension | 52 | 15 | Not defined | Not defined | 48 | 17 |
Peripheral edema | 9 | 0 | 4 | 0 | 15 | 0 |
Cough | 18 | <1 | 9 | 0 | 17 | 2 |
Abdominal pain | 20 | 4 | (11–13) | (0) | 31 | 4 |
Dyspnea | 22 | 3 | 7 | 1 | 21 | 2 |
Decreased weight | 33 | 2 | Not defined | Not defined | 48 | 6 |
Palmer-plantar erthrodysesthesia | 50 | 8 | Not defined | Not defined | 15 | 0 |
Constipation | 25 | <1 | (9–23) | (0) | 37 | 0 |
Pruritus | 8 | 0 | 14 | 0 | 6 | 0 |
Rash | 15 | <1 | 10 | <1 | 17 | 0 |
| Choueiri et al. 2015 [ 54] | CheckMate 025 Trial | Motzer et al. 2015 [ 60, 69, 72] |
The METEOR trial was published in November of 2016, and by April 2016, the FDA had approved cabozantinib as second-line treatment for advanced mRCC after anti-angiogenesis therapy. As to the question of where cabozantinib fits in the sequential treatment paradigm, its superiority to everolimus leaves axitinib as a possible comparator for a future study. Similar to the METEOR trial, the AXIS trial investigated disease refractory to sunitinib. Subgroup and post hoc analyses of the AXIS trial revealed that the patients whom had been treated with sunitinib and axitinib sequentially had a mPFS of 4.8 months and an ORR of 11% [
41,
55]. Considering the 9.1-month mPFS and ORR of 22% observed in this study, cabozantinib could be a marked advancement in the treatment of mRCC. Moreover, the success of cabozantinib serves as a proof-of-principle that the targets (MET and AXL) that were not affected by previous drugs have an in vivo role in mRCC disease.
Lenvatinib in combination with everolimus
One month after the FDA announcement of approval of cabozantinib for mRCC, lenvatinib was approved for the treatment of patients with advanced mRCC in combination with everolimus following disease resistance to TKIs. Lenvatinib is an oral, multi-target TKI of VEGFR1-3, FGFR1-4, PDGFRα, RET, and KIT [
56]. First established as a therapy for differentiated thyroid cancer, lenvatinib has had a favorable antitumor profile with acceptable toxicities in a multitude of solid tumors in both phase I and II trials [
57,
58]. The mechanism by which tumors develop VEGF resistance and develop compensatory angiogenesis pathways provides the rationale for studies on drugs with multiple targets [
59]. Prior in vivo studies using mouse xenografts of human RCC showed a reduction in tumor volume with a lenvatinib and everolimus combination [
60]. Further, in vitro binding studies reveal a highly specific binding site to the receptor kinase domain, suggesting possible limited toxic effects [
61]. To this end, lenvatinib had been identified as a candidate for clinical studies in patients with advanced mRCC refractory to first-line agents.
The role of lenvatinib in the treatment of mRCC has been studied in a randomized, phase II, open-label, multicenter trial, which enrolled 153 patients with mRCC that progressed on first-line VEGF-directed therapy [
62]. Patients were stratified in a 1:1:1 ratio and received lenvatinib, everolimus, or combination therapy with a primary endpoint of mPFS. Lenvatinib plus everolimus significantly prolonged mPFS compared to everolimus, 14.6 to 5.5 months, but not to single-agent lenvatinib, 7.4 months. Moreover, OS was increased in the group receiving the dual-therapy compared to everolimus alone, although not statistically significant. Single-agent lenvatinib significantly prolonged mPFS compared to everolimus as well. However, the size of the benefit of the combination therapy as compared to the benefit of single-agent lenvatinib suggests that efficacy was most robust with the combination therapy [
63]. The design of this three-armed study not only provides objective clinical data but also presents an emerging concept in mRCC therapy that combination drugs targeting multiple pathways (in this case VEGF and mTOR) could simultaneously inhibit two critical independent pathways synergistically and can potentially prevent resistance to single-agent therapy [
62].
The toxicity profile of the combination therapy was consistent with the known toxicities of each individual agent (Table
1). Expectedly, the combination therapy exhibited more frequent adverse events than either single therapy. These most common grade 3 and 4 treatment-emergent adverse events from the dual-therapy patient group include constipation (37%), diarrhea (20%), fatigue (14%), and hypertension (14%). The increased likelihood of toxicity is an appreciable concern and should be considered by clinicians when deciding upon second-line therapy.
Immunotherapy with PD-1 and PD-L1 inhibitors
mRCC is highly immunogenic. Neoplastic cells evade immune cell surveillance allowing for uninhibited and unregulated cell growth. The novel principle underlying immunotherapy entails the activation of the endogenous immune system to target cancer at the cellular level and enable checkpoint inhibition [
64]. Immunotherapy agents at work in mRCC have a role in two-principle immune signaling mechanisms: (1) cytotoxic T-lymphocyte-associated antigen 4 (CTLA-4) and (2) programmed death receptor-1 (PD-1). These two receptors are negative immune regulators imperative to preventing autoimmunity and are endogenous signals for suppression of lymphocytes and natural killer (NK) cells. Although the ability of cancer cells to suppress the immune system is maladaptive, researchers have been able to use this facet of cancer biology as a target for drug development [
65].
The targeted therapies reviewed above and immunotherapies have vastly different mechanisms of action; however, both treatment modalities show durable responses in mRCC tumor regression. Interestingly, studies have shown that targeted therapies are able to dramatically enhance immune cell function as well. Together, this creates a paradigm whereby immunotherapy and targeted therapy can be employed concomitantly with a synergistic effect [
66]. Herein, we describe the immunotherapies approved by the FDA for mRCC including the newly approved agent nivolumab and highlight the promising data that led to nivolumab’s approval.
PD-1 is a cell surface glycoprotein expressed on T-lymphocytes, B-lymphocytes, macrophages, and NK cells. Ligand binding of the PD-1 receptor inhibits the effector phase of T cell activation and thereby serves as a potent immune checkpoint receptor [
6]. The pathogenesis of mRCC cells and their immunomodulatory effect stems from the interface between malignant cells and the PD-1 receptor. The natural ligand to PD-1 is the PD-1 ligand (PD-L1), which is expressed on antigen-presenting cells. Importantly, populations of mRCC cells have also been shown to express PD-L1, and this mimicry can result in not only unregulated tumor cell growth but also apoptosis of antigen-specific T cells [
11,
67]. This interplay between malignant cells and immune cells creates a tumor microenvironment with an active yet functionally impaired immune system. Studies have gone on to show that the degree of PD-L1 expression on mRCC cells directly correlates with aggressive pathologic features including advanced TNM staging, tumor size, higher nuclear grade, coagulative necrosis, increased disease progression, cancer-specific death, and overall mortality [
68].
Nivolumab is a fully humanized immunoglobulin G4 PD-1 immune checkpoint inhibitor antibody that selectively blocks the receptor activation of PD-L1 and PD-L2. Ultimately, nivolumab enhances T cell function which results in antitumor activity [
69]. Nivolumab has changed the landscape for multiple solid and liquid tumors and currently holds FDA approval for the treatment of melanoma, squamous non-small cell lung cancer, and classical Hodgkin’s lymphoma as well as mRCC. The CheckMate016 phase I trial included patients with mRCC and was first presented at ASCO 2015. This study demonstrated that a combination of nivolumab and ipilimumab exhibited a durable antitumor effect with a manageable safety profile [
70]. A recently reported 5-year follow-up investigation of the phase I participants revealed a 34% 5-year survival rate for mRCC patients whom failed prior anti-angiogenesis therapy and then placed on maintenance nivolumab [
71].
Building on these promising results, investigators recruited 168 patients with histological confirmation of mRCC for a phase II study. The patient had received prior treatment with either a VEGF TKI or VEGF monoclonal antibody and suffered from progression of disease. Patients were stratified to receive varying doses of nivolumab as a single-agent therapy. The study successfully demonstrated nivolumab to prolong mPFS (2.7–4.7 months), ORR (20–22%), and mOS (18.2–25.5). Together, the results of the study suggested promising antitumor activity while exhibiting minimal systemic toxicities [
69].
The abovementioned results were encouraging as a proof-of-principle validating immunotherapy as a treatment option for mRCC. CheckMate 025 trial was the first phase III randomized controlled trial examining nivolumab in advanced RCC. Nivolumab was compared to everolimus in disease refractory to VEGFR targeted therapy. Eight hundred twenty-one patients were stratified to receive either nivolumab or everolimus with a primary endpoint of OS. Secondary endpoints included ORR and safety. The results from this study were remarkable and ultimately led to FDA approval of the drug. mOS was 25.0 versus 19.6 months for nivolumab and everolimus, respectively, and met the predetermined criteria for significance. The hazard ratio (HR) for nivolumab met superiority over everolimus (HR 0.73,
p < 0.0148). Further significant findings indicating nivolumab superiority included an ORR of 25 versus 5% and an OR of 5.98. There was no difference in mPFS between the two agents. Moreover, the safety and tolerability profiles favored nivolumab over everolimus (Table
1). Nineteen percent of the nivolumab-treated patients reported grade 3 or 4 treatment-related adverse effects compared to 37% of the patients whom had received everolimus. Of note, the durable responses seen by nivolumab were irrespective of MSKCC prognostic score, number of previous anti-angiogenic therapies, and PD-L1 expression [
72]. Data extrapolated from the CheckMate 025 trial analyzed the role of nivolumab treatment after progression, an emerging concept in clinical oncology [
73]. With regard to the patients in the study treated with nivolumab, 171 of the enrolled patients were treated beyond progression of disease. The patients treated beyond progression experienced a mOS of 28.1 months compared to 15.0 months for the group not treated after progression (
p < 0.001) [
74]. These data are suggestive that the immune response of nivolumab may be delayed, and further studies are necessary to fully elucidate the timing of the clinical effect of the drug.
A 2016 study building off the results from the CheckMate 025 trial compared health-related quality of life (HRQoL) for patients in the treatment groups of this trial. Nivolumab was associated with an improved HRQoL compared to everolimus in this study population [
75]. Although there was no significant difference in the mPFS observed, ad hoc sensitivity analysis of mPFS in the patients whom had not progressed or died within six months of treatment revealed a delay in progression on nivolumab that achieved statistical significance. The landmark CheckMate 025 trial and subsequent studies have resulted in labeling nivolumab as the leading monotherapy for second-line therapy for those who fail VEGFR-targeting therapies [
76].
More recent studies have shown that the immunomodulatory effect of nivolumab in the mRCC tumor microenvironment is expansive [
77]. In an elegantly designed study, baseline and on-treatment biopsies were obtained from mRCC patients receiving nivolumab therapy (both treatment-naïve and refractory disease patients). Immunohistochemical analysis of these biopsies demonstrated an increased lymphocytic presence in the nivolumab-treated group, reversal of T cell exhaustion within the tumor microenvironment, upregulation of genes that are hallmarks of the Th1 inflammatory response, and increased tumor trafficking or infiltration of T cells. The investigators also report an increase in expression of genes linked to NK cells, suggesting that the immunomodulatory effect of nivolumab may be augmented with NK cell-directed therapies in the future [
78].