Lung Cancer 84 (2014) 161–167

Contents lists available at ScienceDirect

Lung Cancer journal homepage: www.elsevier.com/locate/lungcan

Phase I/II trial of vorinostat (SAHA) and erlotinib for non-small cell lung cancer (NSCLC) patients with epidermal growth factor receptor (EGFR) mutations after erlotinib progression夽 Noemi Reguart a,∗ , Rafael Rosell b,c , Felipe Cardenal d , Andres F. Cardona e , Dolores Isla f , Ramon Palmero d , Teresa Moran b , Christian Rolfo g , M. Cinta Pallarès h , Amelia Insa i , Enric Carcereny b , Margarita Majem h , Javier De Castro j , Cristina Queralt b , Miguel A. Molina c , Miquel Taron b,c a

Medical Oncology Department, Hospital Clinic Barcelona-ICMHO, Barcelona, Spain Medical Oncology Department, Hospital Germans Trías i Pujol, Institut Català d’Oncologia-ICO, Badalona, Barcelona, Spain Pangaea Biotech, Hospital Universitario Quirón, Dexeus, Barcelona, Spain d Medical Oncology Department, Centre Sanitari i Universitari de Bellvitge, Institut Català d’Oncologia-ICO, Hospitalet de Llobregat, Barcelona, Spain e Medical Oncology Department, Fundación Santa Fe de Bogotá, Bogotá, Colombia f Medical Oncology Department, Hospital Clínico Lozano Blesa, Zaragoza, Spain g Phase I, Early Clinical Trials Unit, University Hospital of Antwerp-UZA, Belgium h Medical Oncology Department, Hospital de la Santa Creu I Sant Pau, Barcelona, Spain i Medical Oncology Department, Hospital Clínico Universitario, Valencia, Spain j Medical Oncology Department, Hospital La Paz, Madrid, Spain b c

a r t i c l e

i n f o

Article history: Received 28 December 2013 Received in revised form 13 February 2014 Accepted 18 February 2014 Keywords: Vorinostat Erlotinib NSCLC EGFR mutations Histone deacetylase inhibitors TKIs resistance

a b s t r a c t Objectives: Vorinostat or suberoylanilide hydroxamic acid (SAHA) is a novel histone deacetylase inhibitor with demonstrated antiproliferative effects due to drug-induced accumulation of acetylated proteins, including the heat shock protein 90. We prospectively studied the activity of vorinostat plus erlotinib in EGFR-mutated NSCLC patients with progression to tyrosine kinase inhibitors. Patients and methods: We conducted this prospective, non-randomized, multicenter, phase I/II trial to evaluate the maximum tolerated dose, toxicity profile and efficacy of erlotinib and vorinostat. Patients with advanced NSCLC harboring EGFR mutations and progressive disease after a minimum of 12 weeks on erlotinib were included. The maximum tolerated dose of vorinostat plus erlotinib was used as recommended dose for the phase II (RDP2) to assess the efficacy of the combination. The primary end point was progression-free-survival rate at 12 weeks (PFSR12w ). Pre-treatment plasma samples were required to assess T790M resistant mutation. Results: A total of 33 patients were enrolled in the phase I–II trial. The maximum tolerated dose was erlotinib 150 mg p.o., QD, and 400 mg p.o., QD, on days 1–7 and 15–21 in a 28-day cycle. Among the 25 patients treated at the RDP2, the most common toxicities included anemia, fatigue and diarrhea. No responses were observed. PFSR12w was 28% (IC95%: 18.0–37.2); median progression-free survival (PFS) was 8 weeks (IC 95%: 7.43–8.45) and overall survival (OS) 10.3 months (95% CI: 2.4–18.1). Conclusion: Full dose of continuous erlotinib with vorinostat 400 mg p.o., QD on alternative weeks can be safely administered. Still, the combination has no meaningful activity in EGFR-mutated NSCLC patients after TKI progression. © 2014 Elsevier Ireland Ltd. All rights reserved.

1. Introduction 夽 Presented at the 13th World Conference on Lung Cancer, San Francisco July 2009 and the European Multidisciplinary Cancer Congress (ESMO), Berlin, Germany September 2009, Stockholm, Sweden September 2011 and Vienna, Austria October 2012. ∗ Corresponding author at: Medical Oncology Department, ICMHO, Hospital Clinic, C/ Villarroel 170, 08036 Barcelona, Spain. Tel.: +34 932275402; fax: +34 934546520. E-mail addresses: [email protected], [email protected] (N. Reguart). http://dx.doi.org/10.1016/j.lungcan.2014.02.011 0169-5002/© 2014 Elsevier Ireland Ltd. All rights reserved.

Driver mutations at the activation loop of the TK domain of EGFR are present in almost 17% of advanced NSCLC patients [1–3]. Data from prospective phase III trials have demonstrated superiority of TKIs over standard chemotherapy in this molecular selected NSCLC population [4–10]. Unfortunately, the overwhelming majority of responders will inevitably develop acquired resistance [11].

162

N. Reguart et al. / Lung Cancer 84 (2014) 161–167

Table 1 Number of patients with related adverse events (AEs) during phase I portion of the trial (first cycle). Dose level 1 (n = 3)

Dose level 2 (n = 3)

Dose level 3 (n = 8)

Adverse event CTCAE v.3.0

Grade 1–2

Grade 3–4

Grade 1–2

Grade 3–4

Grade 1–2

Grade 3–4

Hematological Anemia Neutropenia Lymphopenia

– – –

– – –

– – –

– – –

3 1 1

– – –

Gastrointestinal Diarrhea Constipation Dysgeusia Nausea Vomiting Xerostomia Dyspepsia Mucositis Thirstiness Dehydration Anorexia Epigastralgia

2 – – – – 2 1 – – – – –

– – – – – – – – – – – –

2 – – 2 – 2 – – 1 – 2 –

– – – – – – – – – – – –

5 1 1 1 3 2 – – – 1 4 2

1a – – 1 – – – 1 – – – –

Skin Cheilitis Hand erythema Rash Pruritus/itching Dry skin Paronychia

– 1 1 – – 1

– – – – – –

1 – 1 – – –

– – 1 – – –

– – 4 1 1 –

– – – – – –

Metabolic/laboratory SGOT (AST) SGPT (ALT) Hyperbilirubinemia Hyperglycemia Hyperkalemia Creatinine

1 – 1 – – –

– – – – – –

– 1 – 1 – 1

– – – – – –

– – 2 – 1 2

– – – – – –

Neurology/endocrine/constitutional Fatigue Fever Somnolence

2 – –

– – –

1 – 1

– – –

5 1 –

1b – –

Abbreviations: CTCAE, common terminology criteria for adverse events; SGOT (AST), serum glutamic oxaloacetic transaminase; SGPT (ALT), serum glutamic pyruvic transaminase; GGT, ␥-glutamyl transpeptidase. a Dose limiting toxicity. b Patient replaced before completing first cycle due to non-treatment compliance.

To date, two major mechanisms of acquired resistance that impair the drug’s inhibitory action have been identified in vivo: a secondary mutation at exon 20 (T790M), which accounts for ∼50% of resistant tumors [12–14], and amplification of the c-Met oncogene (∼10–20%) [15,16]. Several strategies directed against these mechanisms of acquired resistance have been tested so far [17–21]; even so, the standard treatment of advanced EGFR mutated patients after TKI remains to be established. One potential strategy for overcoming resistance is to target EGFR for degradation. One of the most relevant proteins involved in this process is the heat shock protein (Hsp) 90, a well-known client of several oncogenes [22,23]. It is known that EGFR relies on Hsp90 chaperone function for conformational maturation and stability, and the active form of EGFR is more dependent on this chaperone function than its wild-type counterpart [24]. Moreover, Hsp90 maintains the active confirmation of EGFR mutants, preventing ligand-induced down-regulation [25]. Novel Hsp90 inhibitors (IPI-504, AUY922) have shown promising results in EGFR-resistant advanced NSCLC patients [26]. Histone deacetylase inhibitors (HDACI) are compounds that can induce acetylation of Hsp90, disrupting its chaperone function and resulting in tumor cell death [22]. Importantly, treatment of EGFRdependent human lung cancer cell lines with a HDACI induces apoptosis through Hsp90 acetylation and combination with a TKI results in a synergistic tumor cell death [27]. HDAC inhibition

can also induce E-cadherin expression by acetylating the histones recruited by the zinc finger transcriptional repressor (ZEB1) [28] and clinical evidence points to the importance of this epithelial marker as a potential predictor of response to EGFR TKIs [29]. Based on this preclinical evidence, the use of an HDAC inhibitor has been postulated as a reasonable alternative to overcome resistance to TKIs in EGFR-active NSCLC patients. Vorinostat (Zolinza® ; Merck & Co., Inc., Whitehouse Station, NJ, USA), the drug formulation of Suberoylanilide hydroxamic acid or SAHA, is an oral, HDAI with demonstrated antiproliferative effects due to drug-induced accumulation of acetylated proteins [30]. Vorinostat is currently used as monotherapy for primary cutaneous T-cell lymphoma [31], and it has also been tested in combination with chemotherapy without major concerns [32,33]. We conducted this phase I/II trial to demonstrate if the addition of vorinostat to erlotinib could re-sensitize this molecularly selected subset of advanced NSCLC patients with active EGFR and TKI resistance. 2. Patients and methods 2.1. Study design and eligibility criteria This is a phase I/II, non-randomized, open-label, multicenter and investigator-initiated study (ClinicalTrials.gov identifier:

N. Reguart et al. / Lung Cancer 84 (2014) 161–167 Table 2 Frequency of treatment related toxicities occurring in >10% of patients in the phase II portion of the trial (all cycles). Adverse event CTCAE v.3.0

Anemia Leukopenia Diarrhea Fatigue Rash Nausea Vomiting Dry skin Anorexia Xerostomia Conjunctivitis Epigastralgia Neutropenia Mucositis Pneumonitis

163

dose (MTD) was recommended for the phase II (RDP2). Treatment was planned to continue until unacceptable toxicity or disease progression. No intrapatient dose escalation was allowed.

Patients (n = 25) Grade 1–2

Grade 3–4

19 (76%) 3 (12%) 16 (64%) 10 (40%) 12 (48%) 10 (40%) 9 (36%) 8 (32%) 9 (36%) 6 (32%) 5 (20%) 4 (16%) 3 (12%) 3 (12%) –

1 (4%) – 3 (12%) 6 (24%) – 1 (4%) – – 3 (12%) – – – – 1 (4%) 1(4%)

NCT00503971) conducted in Spain under the auspices of the Spanish Lung Cancer Group (SLCG). Patients with confirmed NSCLC histology, advanced disease and EGFR mutations in exons 19 or 21, who had been treated with full doses of erlotinib for a minimum of 3 months were eligible for the study. All patients included in the phase I/II trial were ≥18 years, had an Eastern Cooperative Oncology Group (ECOG) performance status 0–2, measurable disease as defined by the presence of at least 1 lesion measured in one dimension using RECIST (response evaluation criteria in solid tumors) 1.0 and an adequate bone marrow, hepatic and renal function. Patients with asymptomatic central nervous system metastasis who were at least 4 weeks from the time of local therapy were also eligible. Patients with history of hypersensitivity or intolerance to erlotinib who were unable to receive maximum doses were excluded. All patients were required to give written informed consent before enrollment. The Institutional Review Board at each hospital approved the study protocol and the informed consent. The study was conducted in accordance with the principles of the Declaration of Helsinki. 2.2. Drug administration and dose escalation The phase I portion of the study aimed to maintain a continuous full dose of erlotinib 150 mg orally (p.o.) in a daily administration (QD) and escalating doses of vorinostat p.o. at three dose levels: 300 mg QD 7 days every 21 days, 400 mg QD 7 days every 21 days, and 400 mg QD. 7 days every other week. In the event that dose level 1 was not tolerated (DLT), erlotinib dose would be reduced to 100 mg and proceed with dose escalations. Maximum tolerated

2.3. Dose escalation and modification In the phase I, a classic 3 + 3 dose escalation method with 3 patients treated initially at each dose level was used. If none of the first 3 patients experienced DLT, the next dose level was started. If one DLT occurred, then 3 additional patients were treated at the same level. Dose escalation only continued if DLT were observed in no more than 1 patient in the expanded cohort. RDP2 was defined as the highest dose level in which DLT occurred in less than or equal to 1 in 6 patients. The Principal Investigator and members of the Steering Committee took decisions regarding DLT, dose escalation or de-escalation and MTD. A DLT was defined as any of the following during the first cycle of treatment at each level: treatment-related grade 5 hematologic toxicity, grade 4 neutropenia lasting more than 6 days or febrile neutropenia, grade 4 thrombocytopenia or any grade ≥3 non-hematologic toxicities that, in the opinion of the investigator, occurred in the setting of adequate/maximal compliance to conventional medical intervention. 2.4. Evaluation of safety and tolerability In phase I, the RP2D was determined through evaluation of DLTs, as described above. Safety and tolerability were assessed using standard methodology throughout the trial and the 28-day period following the final administration of the combination. Patients were assessed throughout the study for toxicity on the basis of the National Cancer Institute Common Toxicity Criteria for Adverse Events (CTCAE) version 3.0. Baseline assessments were performed the week prior to recruitment and included clinical records, physical evaluation, blood count, chemistry panel and urinalysis. This assessment was repeated weekly during the phase I portion of the trial. 2.5. Efficacy assessments All phase I patients treated at the RDP2 and patients included in the phase II were assessed for efficacy. For tumor response evaluation, a thoracic-abdominal computerized tomography (CT-Scan) was done prior to start of treatment, at cycle 3 and every 2 cycles thereafter until withdrawal from the study or when progressive disease was suspected. Brain CT-Scan and bone scintigraphy were required only if disease was suspected at those levels. Objective tumor responses were assessed throughout the trial according to the RECIST 1.0 [34].

Table 3 Dose escalation and drug delivery modification. Dose level

1

2

3

Vorinostat

300 mg p.o., QD, 7 days on 21 d cycle 3 6 (2–12) 0

400 mg p.o., QD, 7 days on 21 d cycle 3 5 (4–10) 1/0

400 mg p.o, QD, 7 days every other week 25b 2 (1–6) 1/2

1

1

3/2

99/100

98/100

90/92

0

0

1

Patients included Cycles, median (range) Number patients dose reduced V/E (n) Number patients dose delayed V/E (n) Mean dose intensitya V/E (%) DLT

Abbreviations: V, vorinostat; E, erlotinib; p.o., orally; QD, quaque die, daily administration; DLT, dose-limiting toxicity. a Dose intensity defined as the quantity of drug delivered with regards the quantity specified by protocol. b Includes all patients treated at the RDP2

164

N. Reguart et al. / Lung Cancer 84 (2014) 161–167

2.6. Mutational analysis All patients included in this trial were genetically tested EGFR mutations in tissue sample as a part of a Spanish Lung Cancer Group (SLCG) program. Blood sample prior to study was only required for patients included in phase II to assess sensitizing and T790M resistant mutation in serum [4].

2.7. Statistical considerations As the primary purpose of the phase I portion of this trial was to evaluate safety and define the RP2D of the combination, no specific statistical hypothesis was planned. The sample size for phase I was determined based on toxicity, and it was expected that approximately 18 patients would be enrolled. For the phase II portion of the trial, the primary end-point was progression-free-survival rate at 12 weeks (PFSR12w ), defined as proportion of patients alive without disease progression at 12 weeks relative to all patients enrolled. We consider treatment to be efficacious if PFSR12w was ≥60% (alpha of 0.05 and a power of 80%). In the phase II, a total number of 38 patients were estimated for recruitment, with an estimated drop-out of about 10% using Fleming’s procedure [35]. Twenty-five patients treated at the RDP2 were included and evaluable for efficacy. At this point the study was stopped prematurely because it did not meet the goal required for study continuation. Secondary end-points were median progression free survival (PFS) and overall survival (OS). PFS was defined as time from first treatment until progression or death from any cause; OS was defined as time from study inclusion until death.

3. Results Between January 2008 and November 2010, a total of 33 patients were enrolled in the phase I–II trial.

3.1. Safety Fourteen patients were treated with escalation doses of vorinostat. Overall treatment was well tolerated and the majority of AEs observed were grade 1/2 (Table 1). No DLTs were observed at first-second level, rash being the only grade 3 AE reported. A total of 8 patients were included at the third level. Two patients were replaced following the investigators’ decision due to treatment non-compliance. Of those, one developed grade 4 asthenia, however as this patient was off-study it was not considered a DLT. Among the 6 evaluable patients included at the third level, there was only 1 DLT: grade 3 diarrhea. Other Grade 3 non-DLTs reported at this level were 2 episodes of grade 3 mucositis and nausea. Thus the final RDP2 was erlotinib 150 mg p.o., QD and vorinostat 400 mg p.o., QD, on days 1–7 and 15–21 every 28 days. For the phase II, 25 patients were evaluated, including the 6 treated in the phase I at the RDP2. Table 2 summarizes the most common toxicities related to the drug combination. Overall, anemia, diarrhea and fatigue were the most frequent toxicities, experienced by 20 (80%), 19 (76%) and 16 (64%) patients, respectively. The most common grade 3–4 toxicities were fatigue (24%), anorexia (12%) and diarrhea (12%). Three patients stopped treatment due to toxicity: 2 patients had grade 3 fatigue after 2 and 5 cycles, and the other experienced grade 3 pneumonitis related to erlotinib after 16 cycles. Other causes of treatment discontinuation were progressive disease (18 patients), death (1 patient) and investigators’ decision (3 patients). Dose escalation and drug delivery modification for vorinostat and erlotinib are listed in Table 3.

Table 4 Patients’ baseline characteristics. Total n = 25 Median age years (range) Median time from diagnosis, months (range)

60 (42–77) 26.9(13.7–69.2)

Gender n (%) Male Female

9 (36) 16 (64)

ECOG n (%) 0 1 2

8(32) 15(60) 2(8)

Stage IIIB IV

1(4) 24(96)

Smoking habits n (%) Current Former •>5 years •1–5 years •