Exisulind is a well-tolerated11
orally available pro-apoptotic agent6
with preclinical activity in NSCLC.8–10
Vinorelbine is the only drug shown to improve quality of life and survival as a single-agent in the first-line treatment of elderly patients with advanced NSCLC.4
Therefore, it was reasonable to assess these agents together.
Treatment in our phase I population yielded an MTD of 25 mg/m2/week of intravenous vinorelbine with 125 mg of oral exisulind twice daily continuously on a 28-day cycle. DLTs included constipation, diarrhea and febrile neutropenia. The phase II median TTP was 4.7 months (20.4 weeks) and the phase II median OS was 9.6 months (41.7 weeks). Our phase II primary endpoint was met, in that the median TTP exceeded 13 weeks. Neutropenia was the predominant toxicity, although neutropenic fever was rare. The median age of our patients was 78 years.
This is the first clinical study of exisulind with vinorelbine. Four other phase II studies combining exisulind with cytotoxic agents for the treatment of advanced NSCLC have been reported. Masters17
evaluated first-line exisulind combined with carboplatin/gemcitabine and carboplatin/docetaxel, respectively. Hoang12
assessed second-line exisulind combined with gemcitabine and docetaxel, respectively. The results of these four studies were similar to what we would expect with their respective cytotoxic agents alone. In each, exisulind did not worsen outcomes or exacerbate toxicity. These studies’ findings are in line with others in which a targeted agent, such as erlotinib,16
or a matrix metalloproteinase inhibitor,17
was combined with cytotoxic agents for the treatment of advanced NSCLC. The notable exception to this record is the prolongation of survival resulting from the addition of bevacizumab to cytotoxic agents.22, 23
Our efficacy results compare favorably with those of prior single-agent vinorelbine studies treating advanced NSCLC in the elderly. In a phase II study by Gridelli, patients at least 70 years of age with stage IIIB-IV NSCLC and an ECOG PS ≤ 2 were treated with first-line vinorelbine (30 mg/m2
The median age was 73 years and median ECOG PS was 2 (22/43 patients). Median TTP was 11 weeks and median OS was 36 weeks. A fair assertion when comparing this study with ours is that our patients were “fit elderly” with a median ECOG PS of 1. Nonetheless, the finding in our elderly phase II population of a median TTP of 4.7 months and a median OS of 9.6 months is provocative when compared to single-agent historical control.13
A randomized trial of vinorelbine with and without exisulind would be necessary to identify any clinical benefit in adding exisulind.
It is unclear why the results of our combination study incorporating exisulind yielded more favorable efficacy results than seen with prior exisulind combination studies in advanced NSCLC.12, 15, 18, 19
. Although small in number, our patients were selected from multiple sites in our academic-community network. In addition, there do not appear to be significant discrepancies between our eligibility criteria and those from the other studies.16–19
Our findings are consistent with the prior exisulind studies, in that the addition of exisulind to a cytotoxic regimen did not appear to exacerbate toxicity. Although preclinical evidence to date suggests a lack of potentiation between exisulind and cytotoxics,20
our favorable experience with this combination prompts us to consider the use of pro-apoptotic agents in NSCLC worthy of further study.
The predominant toxicity in our patients was neutropenia attributable to vinorelbine. This was expected.4, 21
Nearly half of phase II patients experienced grade ≥ 3 neutropenia. However, neutropenic fever was uncommon. Prophylactic G-CSF support may have been useful in this setting to avoid treatment delay and dose reduction. Prophylactic pegfilgrastim starting with the first cycle of chemotherapy reduced the rate of neutropenic fever and hospitalizations in a randomized trial of 852 patients at least 65 years old.22
Since the initiation of our study, several organizations23–25
have recommended consideration of prophylactic G-CSF in elderly patients receiving standard-dose myelotoxic chemotherapy for whom neutropenia is expected.
The addition of exisulind to vinorelbine did not worsen toxicity in our patients. This is evidenced by the fact that 40% of our phase II patients completed six cycles of combination treatment. Additionally, in 175 phase II cycles of exisulind as part of combination therapy or as maintenance, exisulind-related toxicities were ≤ grade 2 in all but two cases (grade 3 ALP and grade 3 AST elevation). This profile is similar to that seen in a phase I study by von Stolk of single-agent exisulind in patients aged 18–45 years with FAP and subtotal colectomy.11
Here, there was only one grade ≥ 3 event (elevated alanine aminotransaminase) over approximately 108 cycles. We did not perform pharmacokinetic analyses. Therefore, one may theorize that the von Stolk study found a higher MTD because of an undetected interaction between exisulind and vinorelbine. However, it is also possible that our patients, who had intact colons, had greater resorption of exisulind leading to a prolonged residence time.26, 27
It is noteworthy that among our 30 phase II patients, eight developed grade 3,
and one developed grade 4, dyspnea. Dyspnea is a nearly ubiquitous symptom in this population, in whom pleural effusions, pulmonary infections, pulmonary emboli, atelectasis, emphysema, anemia, cardiovascular disease, and prior treatments with radiotherapy are frequent.28
For example, Gridelli et al. reported a baseline incidence of dyspnea of 67% in their elderly cohort.3
Dyspnea is not a known adverse effect of exisulind, and prior evaluation of vinorelbine in the elderly do not report the incidence of dyspnea during treatment.4, 11, 13
It does not appear that exisulind contributed to the worsening of dyspnea in our patients.
Strengths of this study include its novelty as the first clinical study of vinorelbine with exisulind and its focus on older patients. Additionally, the conduct of this trial in both the academic and community settings may provide less bias in patient selection and a more realistic profile of toxicities and patient outcomes.
There are several limitations to interpreting this study. First, we did not use a uniform method of classifying comorbidities and functional impairment.29–33
We relied on ECOG PS3
in determining eligibility. As has been documented, the use of ECOG PS alone may not be the best way to assess the functionality of the elderly.34
Second, pharmacokinetic analyses were not conducted. Pharmacokinetic studies in elderly cancer patients have provided conflicting results, including reports analyzing vinorelbine.35–37
To our knowledge, no data exist concerning the clearance of oral exisulind in the elderly. However, Sitar found no differences in the clearance of sulindac sulfoxide, a prodrug of exisulind, when comparing elimination data from young, healthy individuals with those older than 65 years taking 150 mg twice a day.38
The type, frequency and severity of our phase II patients’ toxicities were in line with our expectations. This observation implies there is low likelihood of significant pharmacologic interaction.
Third, 30% of our phase II patients presented for first-line chemotherapy after prior surgery, and 37% after prior radiation. This may indicate that our cohort carried a more favorable prognosis and may have contributed to our provocative efficacy outcomes.
Last, most of our patients had a PS of 1, and only two of 30 had a PS of 2. Therefore, our results are confined to the “fit elderly” and may not generalize to a broader geriatric cancer population.
We demonstrate that a first-line regimen of vinorelbine/exisulind was reasonably well tolerated by an older cohort of patients with advanced NSCLC. Median TTP and median OS were prolonged compared to historical control. We conclude that this combination is safe, appears to have activity in patients at least 70 years of age with advanced NSCLC and a PS ≤ 2, and that the use of pro-apoptotic agents in NSCLC warrants further investigation. At present, however, the study sponsor does not plan further clinical development of exisulind.