In this analysis of mRCC patients in the PROCLAIMSM HD IL-2 registry, all IMDC risk groups have median and 2-year survivals following HD IL-2 that are consistent with recent reports of CPI immunotherapy or anti-VEGF targeted therapy for mRCC. All risk categories have improved survival compared with historical cytokine data [11,12,13,14,15] and are consistent with data from a recent prospective study, IL-2 “Select” [5] and other contemporary reports [2,3,4, 6,7,8].
In the high dose IL-2 alone patient group, favorable and intermediate risk patients demonstrate prolonged OS, and many experience years of treatment-free survival following IL-2 therapy. Further, the prolonged OS and 74% 2-year survival for the favorable risk group treated with IL-2 alone is in contrast to the outcome from the randomized trial of combined checkpoint inhibition, in which the favorable risk group had a better 18-mo OS with sunitinib compared with CPI [18]. This contrast demonstrates the nuances and differences between different categories of immunotherapy in the treatment of mRCC that are yet to be sorted out.
Eligibility for HD IL-2 treatment includes physiologic evaluation which may enrich for favorable risk patients, although the percentage of intermediate risk patients is similar to other mRCC treatment reports. Nevertheless, these eligibility criteria might inadvertently select an even better subset of each IMDC risk group for outcome when treated with IL-2 alone or with IL-2 followed by subsequent therapy, yielding durable response and survival. In support of this concept, even poor risk patients meeting IL-2 therapy criteria demonstrated clinical benefit with best clinical responses of CR, PR and SD reported, and 2 year OS of 40–50%.
Among responders in all treatment sequences, median response duration (CR and PR) following HD IL-2 was greater than or equal to 5 years regardless of risk category. However, durable stable disease was observed primarily in favorable risk (median > 5 years) and intermediate risk (median > 3 years) groups, with 10% durable SD among the poor risk group.
The CR rate of 5.4% is consistent with prior reports of IL-2 alone, but less than that of the combination ipilimumab/nivolumab arm of Checkpoint 214 in first line patients [18]. However, it is of note that the CR rate for single agent nivolumab in mRCC in both a large clinical trial experience and a large registry experience is 1 and 1.2%, respectively [20, 21]. Additionally, the CR plus PR rate for HD IL-2 alone is 25%, again consistent with prior reports of IL-2 and possibly allowing for resection of residual disease in PR patients, yielding “surgical CRs”, often with long-term disease-free intervals. The CR + PR rate for nivolumab alone was reported as 25 and 20% [20, 21].
Limitations of registry data
Therapy-specific registries have helped provide real-world data and increased safety data on therapies beyond the initial clinical trials leading to drug approvals, and thus provide more insight into the spectrum of use of therapies. Examples are the International Bone Marrow Transplant Registry (IBMTR), the IMDC, and the initial registries of patients treated with anti-VEGF targeted therapies for mRCC patients off protocol, following regulatory approvals, but before general availability of these medications. The PROCLAIM registry has provided insight into optimization of IL-2 management and patient selection for this therapy with a goal of maximizing benefit from IL-2 [19, 22].
With decades of experience, careful patient selection and explicit treatment eligibility criteria have enhanced the safety and efficacy of IL-2, but such selection may limit generalizability of conclusions. There are limitations to registry data in that it may not be audited or reviewed, relying on investigator reporting, and there are limitations to the amount of data collectable from sites, as well as variations in documentation, which may limit comparability. Additionally, patients may not be enrolled in a consecutive fashion, depending on other treatment options or choices available at various sites.
Combination of HD IL-2 with checkpoint inhibitors, sequential and concurrent treatment
In view of the well-known durability of response (DOR) to HD IL-2 from many reports, and the observation of durable survival of mRCC patients following HD IL-2, particularly in favorable and intermediate risk groups in this report, combination therapy, both in sequence and concurrently, to enhance the complete response rate, DOR and OS is being investigated. Metastatic RCC responders to HD IL-2 have among the most durable survival data of mRCC patients with any treatment [2,3,4,5,6,7,8, 10, 12, 16,17,18]. A series of recent reports suggest the feasibility of combinations, either sequentially or concurrently.
Safety and efficacy of HD IL-2 followed by anti-VEGF targeted therapies for stable or progressive disease has been previously reported from the PROCLAIM registry data, and confirmed in this larger cohort (Additional file 1: Table S1a, b), demonstrating enhanced OS [6]. The increased use of anti-VEGF therapy and CPI therapy as initial treatment has also led to further evaluation of sequence, with these therapies preceding HD IL-2. As presented in the supplemental data patients eligible to receive HD IL-2 following progression on anti-VEGF therapy have outcomes similar to those receiving HD IL-2 alone (Additional file 1: Table S1b).
Buchbinder et al. has published retrospective data demonstrating activity and no additive or unexpected toxicity among melanoma patients treated with HD IL-2 following progression after ipilimumab (Ipi) [23]. More recently, they reported safety and efficacy of HD IL-2 treatment in patients who had previously received anti-PD1 therapy and then progressed, (including both melanoma and mRCC patients) with outcomes similar to patients treated with IL-2 alone as first line therapy [24]. HD IL-2 was both active and safe in patients who had no ongoing immune-related adverse events (iRAEs) other than hypothyroid disease undergoing replacement therapy [24]. A prospective evaluation of this sequence is planned.
With respect to concurrent administration of HD IL-2 and CPI therapy, early feasibility was reported by Prieto et al. in an initial clinical study report and a 7-year follow-up of 36 melanoma patients treated with the combination of Ipi and HD IL-2 [25, 26]. The schedule was dose 1 of Ipi given alone, followed by dosing every 3 weeks, as tolerated, with the combination of one dose of Ipi and then HD IL-2 720,000 units per kilogram (U/kg) every 8 h up to 15 doses, with the first dose of IL-2 given within 24 h of the Ipi dose. Although initially a dose-finding study, 24 of 36 patients were treated at 3 mg/kg of Ipi. The CR rate (calculated among all 36 patients) was 17%, with all CR’s ongoing at the time of the long-term follow-up report and the longest duration being 89+ months in that report [26]. Of interest, grade III/IV iRAEs were 17% among the Ipi/IL-2 patients, in contrast to grade III/IV iRAE rates of 29 and 32% in two parallel Ipi/vaccine trials conducted at the same doses and schedules of Ipi by this group [25, 26]. Also of interest, among the Ipi/IL-2 patients, there was no correlation between response and development of iRAEs, but the numbers are small. The high CR rate and tolerable toxicity as well as the durability of responses observed, strongly suggest that further combination studies of HD IL-2 and CPIs should be evaluated, for the potential of enhancing CR rate, potentially leading to durable responses and enhanced 2 year and overall survival.
Several ongoing investigator initiated trials are evaluating concurrent or rapid sequential use of HD IL-2 and anti-PD1 agents, in both melanoma and mRCC, and are evaluating immune parameters and biomarkers (Additional file 1: Table S2). These studies are ongoing, several are dose-finding for IL-2, and to date no unusual safety signals have been reported. The clinical trial that is furthest along is NCT02964078, in mRCC patients, with concurrent pembrolizumab and IL-2, 600,000 U/kg, but utilizing a different IL-2 schedule, with 5 doses administered over 33 h, similar to a previously published regimen [27]. Preliminary data were reported at the February 2019 annual meeting of the Genitourinary American Society of Clinical Oncology (ASCO) and describe a higher than additive response rate with no prohibitive toxicity [28]. Follow-up is ongoing.