A similar phase I clinical study using ISIS 5132, a first-generation
ASO targeted to the MAP kinase signaling pathway protein RAF1 [77], escalated doses up to
5 mg/kg/d without reaching a maximum tolerated dose [78]. The difference in apparent tolerability
of ISIS 3521 and ISIS 5132 is not clear, and may not even be real, since it may reflect nondrug
related adverse events, or differences in patient population. Nevertheless, based in part on these
observations and on preclinical models demonstrating activity at equivalent dosing, the phase II
dose for ISIS 3521 as well as ISIS 5132 was set to 2 mg/kg/d. Both agents were compared in a
National Cancer Institute of Canada randomized phase II trial in patients with hormone-refractory
prostate cancer [79]. Scheduling for both ASOs was 2 mg/kg/d for 21 days by continuous infusion
followed by a 7-day rest period. Overall, treatment was well tolerated, with fatigue and mild thrombocytopenia as the main
treatment-related adverse events. However, although some patients had
stable disease, no biochemical or objective responses were observed. A following phase I/II combination trial tested ISIS 3521 at
2 mg/kg/d by continuous infusion on days 0–14 with cisplatin at
80 mg/m2(day 1) and gemcitabine at 1000 mg/m2(day 1 and 8) [80] in 55 chemotherapy-naïve
patients with advanced nonsmall cell lung cancer (NSCLC). Sixteen of 48 evaluable patients had a
response (1 complete response and 15 partial responses). The median survival time for the entire
group of 55 patients was 8.9 months and 10.5 months for the 45 patients receiving 2 full cycles
of treatment. Toxicity was moderate but included thrombocytopenia, neutropenia, anemia, fatigue,
dehydration, sepsis, and neutropenic fever. On the basis of this phase II data and another combination study by Yuen et al. [81]
with carboplatin and paclitaxel that had a reported 42% response, two
large randomized phase III trials were conducted as first line treatment in patients with NSCLC. The
first enrolled 600 patients using ISIS 3521 in combination with carboplatin and paclitaxel in
patients with Stage IV NSCLC, and the results were disappointing [82]. No difference was observed
in time to progression or overall survival between groups. There were, however, indications of
antitumor activity, as patients who completed the prescribed course of therapy (6 cycles) receiving
ISIS 3521 had a median survival of 17.4 months compared to 14.3 months in patients who did not
( p 0.048). Negative results were also obtained in the other phase III NSCLC trial testing ISIS
3521 in combination with gemcitabine and cisplatin. Therapy was fairly well tolerated, but median
survival was 10 months in both groups [83].
Several factors may have accounted for the lack of clinical efficacy for ISIS 3521. First, because
measuring target levels in tumors is difficult in this patient population, patients were not screened
and it is unknown whether the target was actually expressed in the majority of patients. Therefore
the target itself may not have been relevant in the studied cohort. Moreover, recent preclinical data
suggests that in NSCLC other PKC isoforms than PKC may be a driving force for cancer cell
survival [84]. Secondly, inhibition of PKC may not produce a large enough effect on tumor growth
and only result in a cytostatic effect, which the study design would not detect. Third, inhibiting a
single molecular target may be insufficient to exert a clinically detectable effect beyond what can
be achieved with combination chemotherapy alone. Finally, since target knockdown was not
assessed, the dose of ASO used may not have been the optimal biological dose and hence not potent
enough to inhibit PKC expression sufficiently.
