Author of

• 11745

  +

  P2.11 - Poster Session 2 - NSCLC Novel Therapies (ID 209)

  • Event: WCLC 2013
  • Type: Poster Session
  • Track: Medical Oncology
  • Presentations: 1
  • Moderators:
  • Coordinates: 10/29/2013, 09:30 - 16:30, Exhibit Hall, Ground Level

  • 11786

    +

    P2.11-041 - Effect of treatment duration on incidence of adverse events (AEs) in a phase III study of crizotinib versus chemotherapy in advanced <em>ALK</em>-positive non-small cell lung cancer (NSCLC) (ID 2900)

    09:30 - 09:30  |  Author(s): A.T. Shaw
    • Abstract

    Background
    In the ongoing global randomized phase III study PROFILE 1007, the oral ALK inhibitor crizotinib improved progression-free survival, response rate, and global quality of life compared with single-agent chemotherapy in patients with advanced, previously treated ALK-positive NSCLC. While hematologic toxicities were more common with chemotherapy, a greater number of non-hematologic toxicities were reported with crizotinib. Several factors may have complicated the comparison of AEs, including longer treatment duration with crizotinib, standard use of prophylactic medications with chemotherapy but not with crizotinib, and a higher rate of continuing treatment beyond RECIST-defined progressive disease with crizotinib. Here we examine whether differences in treatment duration may have impacted the incidences of selected AEs and non-fatal serious AEs (SAEs) observed in this study.

    Methods
    In PROFILE 1007, 172 patients received crizotinib 250 mg orally BID and 171 patients received chemotherapy (pemetrexed 500 mg/m[2] or docetaxel 75 mg/m[2], both IV q3w). AEs were classified and graded using CTCAE v4.0. For this analysis, AEs of interest included those occurring in ≥10% of patients with ≥5% absolute difference between treatment groups, as well as non-fatal SAEs. To evaluate the potential impact of differential treatment duration on these AEs, exposure-adjusted incidence rates were calculated using person-years of exposure (PYE; the sum of the individual person-years at risk for a particular AE).

    Results
    Median treatment duration was almost three-fold longer with crizotinib (31 weeks) versus chemotherapy (12 weeks). Incidences and exposure-adjusted incidence rates of AEs of interest are shown in the table below. After accounting for treatment duration, the rates of nausea, constipation, elevated transaminases, edema, upper respiratory infection, dizziness, pulmonary embolism, hypokalemia, and non-fatal SAEs were comparable between the crizotinib and chemotherapy groups. The rates of diarrhea, vision disorder, vomiting, dysgeusia, and syncope were significantly higher with crizotinib. Exposure-adjusted incidence rates for other AEs, as well as additional analyses to account for treatment duration will be presented.

    	Crizotinib (n=172)			Chemotherapy (n=171)
    AE	Incidence (%)	Incidence rate/1000 PYE		Incidence (%)	Incidence rate/1000 PYE	P value[a]
    Diarrhea	60	1848		19	592	<0.0001
    Vision disorder[b]	60	2449		9	268	<0.0001
    Nausea	55	1577		37	1527	0.842
    Vomiting	47	1083		18	561	0.001
    Constipation	42	832		23	789	0.789
    Elevated transaminases[b]	38	716		15	470	0.056
    Edema[b]	31	568		16	473	0.430
    Dysgeusia	26	463		9	269	0.045
    Upper respiratory infection[b]	26	419		13	401	0.865
    Dizziness[b]	22	346		8	236	0.195
    Pulmonary embolism[b]	6	78		2	62	0.686
    Hypokalemia	5	70		2	63	0.846
    Syncope	3	39		0	0	0.025
    Non-fatal SAEs	29	440		21	640	0.105

    [a]For differences in incidence rates, two‑sided.[b]Clustered term.

    Conclusion
    This initial analysis suggests that differences in treatment duration between two drugs may have a significant impact when comparing AE profiles. Depending on the safety profiles of the two drugs, analyses to account for treatment duration may be appropriate when a large disparity has been observed.