EURURO Vol. 72 No. 6

(

= 0.0475). There was no evidence of an interaction for

other subgroups.

3.7.

ORR for each subgroup

ORR favored nivolumab over everolimus in all subgroups,

most notably in patients with poor MSKCC risk, aged at least

65 yr, one site of metastasis, lung metastases, prior

pazopanib therapy, and two prior antiangiogenic therapies

( Fig. 1

B). There was no evidence of an interaction between

treatment and each subgroup with ORR.

3.8.

Safety for each subgroup

The incidence of grade 3 or 4 TRAEs was lower in the

nivolumab arm compared

[4_TD$DIFF]

with the everolimus arm across

subgroups

( Table 2

). The incidence of grade 3 or 4 TRAEs in

the nivolumab arm was half or less than half the incidence

observed with everolimus in the following subgroups:

favorable MSKCC risk (15% vs 40%); favorable IMDC risk

(20% vs 41%); intermediate IMDC risk (17% vs 36%); at least

65 yr of age (16% vs 39%); one site of metastasis (15% vs

33%); lung metastases (17% vs 37%); prior interleukin-2

therapy (19% vs 38%); at least 6 mo on first-line therapy

(17% vs 37%); and two prior antiangiogenic therapies (17%

vs 36%;

Table 2

). There was no evidence of an interaction

between treatment and each subgroup with any TRAE.

Discussion

With a minimum follow-up of 14 mo in previously treated

patients with aRCC, OS and ORR favored nivolumab over

everolimus for multiple subgroups. Within the nivolumab

arm, ORR for most subgroups was similar and consistent

with the overall ORR reported previously

[7]

Median OS was greater and mortality rate was lower

with nivolumab than with everolimus in all MSKCC risk

groups, with the largest difference in patients with poor

risk. The small number of events and short duration of

follow-up in the favorable risk group limited the ability to

observe robust OS differences between arms. The large

difference for poor-risk patients suggests that further

investigation of the characteristics of these patients, such

as tumor biology, is needed to better understand this

finding. One explanation, yet to be formally analyzed, is the

potential presence of a higher mutational load in poor-risk

patients, a phenomenon that in some cases has been

correlated with better efficacy of PD-1 inhibitors. This is

Table 2 – Summary of treatment-related adverse events for patients within each subgroup

Subgroup

Treatment-related adverse events,

(%)

Nivolumab

Everolimus

Any grade

Grade 3–4

Any grade

Grade 3–4

All treated patients

319 (79)

76 (19)

349 (88)

145 (37)

MSKCC risk score

Favorable

116 (85)

21 (15)

131 (93)

56 (40)

Intermediate

143 (74)

36 (19)

160 (87)

63 (34)

Poor

60 (78)

19 (25)

58 (80)

26 (36)

IMDC risk score

Favorable

46 (84)

11 (20)

67 (99)

28 (41)

Intermediate

189 (78)

41 (17)

202 (87)

83 (36)

Poor

71 (76)

21 (22)

64 (80)

26 (33)

Not reported

13 (81)

3 (19)

16 (100)

8 (50)

Age group

65 yr

200 (79)

51 (20)

199 (86)

81 (35)

65 yr

119 (78)

25 (16)

150 (90)

64 (39)

Number of sites of metastases

57 (85)

10 (15)

62 (90)

23 (33)

262 (77)

66 (19)

287 (88)

122 (37)

Site of metastases

Bone

51 (67)

13 (17)

50 (75)

13 (19)

Liver

78 (80)

22 (22)

72 (85)

31 (36)

Lung

215 (78)

46 (17)

232 (88)

97 (37)

Prior therapy

Sunitinib

197 (78)

46 (18)

220 (88)

89 (35)

Pazopanib

103 (82)

24 (19)

114 (89)

45 (35)

Interleukin-2

37 (88)

8 (19)

30 (88)

13 (38)

Time on first-line therapy

6 mo

85 (79)

24 (22)

109 (88)

44 (35)

6 mo

234 (78)

52 (17)

240 (88)

101 (37)

Prior antiangiogenic therapies

240 (76)

61 (19)

264 (87)

111 (37)

76 (85)

15 (17)

85 (90)

34 (36)

IMDC = International Metastatic Renal Cell Carcinoma Database Consortium; MSKCC = Memorial Sloan Kettering Cancer Center.

Two grade 5 events (septic shock and acute bowel ischemia).

One grade 5 event.

E U R O P E A N U R O L O G Y 7 2 ( 2 0 1 7 ) 9 6 2 – 9 7 1

969