- Email: [email protected]
Current Readings: Window-of-Opportunity Trials for Thoracic Malignancies
CURRENT READINGS
Current Readings: Window-of-Opportunity Trials for Thoracic Malignancies Anne S. Tsao, MD Recent major advances in metastatic non–small cell lung cancer have occurred with the identification of molecular biomarker targets and administration of novel agents with resulting improvement in clinical outcomes. In the early-stage setting, personalized therapy with novel agents and molecular profiling are being incorporated into neoadjuvant “window-of-opportunity” trials. These important studies enable biomarker research and an expedited analysis of the efficacy of the targeted agent. However, there are significant limitations to window-of-opportunity trials. The aim of this article is to review the current window-of-opportunity trials of neoadjuvant targeted agents for thoracic malignancies, discuss the benefits and limitations of these trials, and propose more optimal alternative trial end points. Neoadjuvant trials of resectable non–small cell lung cancer and mesothelioma that are ongoing or under development and relevant to thoracic surgeons are also discussed. The success of these trials will depend on a collaborative multidisciplinary effort, especially from the field of thoracic surgery. Semin Thoracic Surg 26:323–330 I 2014 Elsevier Inc. All rights reserved. Keywords: NSCLC, window-of-opportunity trials, targeted agents, neoadjuvant, thoracic malignancies INTRODUCTION Neoadjuvant chemotherapy is a common approach to managing resectable local or regionally advanced non–small cell lung cancer (NSCLC). Most neoadjuvant trials have used 4 cycles of platinum-doublet chemotherapy and, as a whole, have shown a 5% absolute survival benefit at 5 years.1 The advantages and disadvantages to using neoadjuvant chemotherapy are listed in Table 1. Clinical trials using neoadjuvant chemotherapy or chemoradiotherapy have identified prognostic factors for improved survival after tumor resection.2-5 Patients with superior survival outcomes include those who achieve a response to chemotherapy, whose disease is downstaged from N2, who have single-station N2 rather than multistation N2 disease, who achieve R0 resection, and who have a complete pathologic response.2,4 Department of Thoracic & Head and Neck Medical Oncology, The University of Texas MD Anderson Cancer Center, Houston, Texas Dr Tsao reports receiving consulting fees from, Roche, Genentech, Novartis, Boehringer-Ingelheim, Medimmune, and Lilly. Address reprint requests to Anne S. Tsao, MD, Department of Thoracic & Head and Neck Medical Oncology, The University of Texas MD Anderson Cancer Center, 1515 Holcombe Blvd., Unit 432, Houston, TX 77030. E-mail: [email protected]
1043-0679/$-see front matter ª 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1053/j.semtcvs.2014.12.005
Thus, it can be extrapolated that better systemic therapies may lead to increased downstaging and superior clinical outcomes. In the era of personalized therapy with novel agents and molecular profiling, designing studies—such as “window-of-opportunity” trials—that enable biomarker research and expedited analysis of whether the agents are efficacious are critically needed. This review article summarizes the current literature on 3 thoracic window-of-opportunity studies with novel targeted agents to illustrate the evolution of this type of trial design. The advantages and disadvantages of novel agent window-ofopportunity trials and alternative trial end points are discussed. Ongoing neoadjuvant studies with novel agents relevant to the thoracic surgical community are summarized. Window-of-opportunity studies with multidisciplinary effort, especially surgical support, are a necessary and vital component to move the field of thoracic oncology forward. PHASE II STUDY OF PREOPERATIVE GEFITINIB IN CLINICAL STAGE I NON– SMALL CELL LUNG CANCER Lara-Guerra H, Waddell T, Salvarrey M, et al: J Clin Oncol 27:6229-6236, 2009 Lara-Guerra et al6 conducted a preoperative window-of-opportunity trial (Fig. 1) using gefitinib, an epidermal growth factor receptor (EGFR) tyrosine 323
WINDOW-OF-OPPORTUNITY TRIALS FOR THORACIC MALIGNANCIES Table 1. Hypothetical Implications of Neoadjuvant Therapy Advantages
Disadvantages
Gives patients time to quit using tobacco Eliminates micrometastatic disease earlier
Delays definitive procedure Interferes with surgery owing to the toxicity of chemotherapy Creates staging ambiguity
Allows higher dose chemotherapy (patients tolerate chemotherapy better before rather than after major surgery) May lead to downstaging of disease Provides prognostic value (assessment of chemosensitivity)
kinase inhibitor, and demonstrated that this type of trial design is feasible and safe for patients with earlystage resectable non–small cell lung cancer (NSCLC). In this single-arm phase II study, Lara-Guerra et al6 enrolled 36 patients with stage I resectable NSCLC and treated them with preoperative gefitinib (250 mg/d) for up to 28 days. Radiographic imaging results as determined using standard response evaluation criteria in solid tumors (RECIST) were obtained before and after preoperative gefitinib therapy. After 24-48 hours from the last dose of neoadjuvant gefitinib, patients underwent mediastinoscopy and surgical resection. The primary end point of the trial was overall response rate as determined by RECIST, and the secondary end points included the overall survival and progression-free survival rates. The translational studies included analysis of tumor tissue for EGFR by immunohistochemistry and fluorescent in situ hybridization, EGFR gene mutation, and serum levels for tumor growth factor α. Among the 35 evaluable patients, 11% had a partial response, 81% stable disease, and 8% progression of disease as determined by RECIST after 28
Increases the risk of postoperative complications
days of gefitinib treatment. Tumors shrank in 15 (43%) patients, grew in 15 patients (43%), and did not change in size in 5 (14%) patients. The only biomarker that predicted a clinical partial response was the presence of a sensitizing EGFR mutation (deletion exon 19 or L858 mutation). Among the patients with a KRAS mutation, there were no responders. No other biomarkers were predictive or prognostic, although in one case a patient with a high EGFR gene copy number as determined by fluorescent in situ hybridization and no EGFR mutation also demonstrated a partial response.
MOLECULAR CHARACTERISTICS PREDICT CLINICAL OUTCOMES: PROSPECTIVE TRIAL CORRELATING RESPONSE TO THE EPIDERMAL GROWTH FACTOR RECEPTOR TYROSINE KINASE INHIBITOR GEFITINIB WITH THE PRESENCE OF SENSITIZING MUTATIONS IN THE TYROSINE BINDING DOMAIN OF THE EGFR GENE Rizvi N, Rusch V, Pao W, et al: Clin Cancer Res 17:3500-3506, 2011
Figure 1. Phase II trial schema of preoperative gefitinib for patients with stage I resectable NSCLC.6 FISH, fluorescent in situ hybridization; FNA, fine needle aspiration; ORR, overall response rate; OS, overall survival; PFS, progression-free survival; TGFα, transforming growth factor α.
324
Seminars in Thoracic and Cardiovascular Surgery Volume 26, Number 4
WINDOW-OF-OPPORTUNITY TRIALS FOR THORACIC MALIGNANCIES Rizvi et al7 (Fig. 2) took non–small cell lung cancer (NSCLC) window-of-opportunity trials to the next step in this study that selected patients by clinical characteristics to enhance the likelihood of enrolling patients who had epidermal growth factor receptor (EGFR) mutations. This study verified that the presence of a sensitizing EGFR mutation correlated with a radiographic response. However, there was no difference in survival outcomes, which was likely owing to the short duration of neoadjuvant gefitinib therapy and the inability of many patients to tolerate adjuvant maintenance gefitinib therapy for 2 years. Rizvi et al7 enrolled patients with stage I-II NSCLC who had a o15 pack-year smoking history, a bronchioloalveolar component on histologic assessment, or both to a single-arm trial. The clinical selection was designed to enrich the population with sensitizing EGFR mutations, which at the time of the trial design had just been identified to be predictive for dramatic responses to EGFR tyrosine kinase inhibitors.8,9 In this trial, patients with NSCLC were treated with gefitinib therapy (250 mg/d) for up to 21 days and 48 hours later underwent surgical resection and adjuvant chemotherapy if appropriate. Radiographic images were obtained before and after gefitinib therapy, and the tumors were measured using the World Health Organization (WHO) bidimensional measurement criteria. Patients who had a radiographic response or identification of a sensitizing EGFR mutation were offered adjuvant maintenance gefitinib (250 mg/d) for up to 2 years. Patients who did not have an EGFR mutation and did not have a radiographic response to gefitinib were observed after resection. The primary end point was to correlate the WHO bidimensional radiographic response to EGFR mutation status. Secondary end points included overall survival and progression-free survival rates.
A total of 50 patients were enrolled (39 women and 11 men). Among them, 36% of patients were never smokers and 46% were r15 pack-year former smokers. There were 9 patients who had bronchioloalveolar features on histologic analysis. In total, 21 patients had a radiographic response to neoadjuvant gefitinib, all of whom had undergone surgical resection, whereas among the 29 non-responders, 3 could not undergo resection owing to upstaging. Of the 47 patients who underwent surgery, 6 also received adjuvant chemotherapy. Postoperatively, adjuvant gefitinib maintenance therapy was offered to the 21 responding patients, which was initiated on postoperative day 7. Postoperative complications included 2 cases of pneumonitis, 2 of atrial fibrillation, 1 of supraventricular arrhythmia, 3 of air leaks, and 1 of intraoperative injury to the right main pulmonary artery. The clinical results included 21 patients with Z25% tumor shrinkage by WHO bidimensional measurements after 21 days of gefitinib therapy. Of the 21 responding patients, 17 had an EGFR mutation. In addition, 5 patients had a KRAS mutation, and none had a radiographic response. There was no correlation between EGFR immunohistochemistry results or gene amplification. In one curious case, the tumor of a patient with a PI3KCA (E542K) mutation and a BRAF (V600E) mutation shrank by 74% in response to gefitinib therapy. In this trial, the median progression-free survival rate was not reached. The analysis was compromised because less than half of the 21 patients eligible for the 2 years of adjuvant gefitinib were able to complete the therapy. The 2-year disease-free survival rate was 95% for patients who received adjuvant gefitinib and 78% for patients who did not. The 2-year disease-free survival rate was 95% for patients with an EGFR mutation and 90% for those who did not.
Figure 2. Trial schema of preoperative gefitinib for EGFR mutation–enriched population of patients with stage I-II NSCLC.7 BAC, bronchioloalveolar carcinoma.
Seminars in Thoracic and Cardiovascular Surgery Volume 26, Number 4
325
WINDOW-OF-OPPORTUNITY TRIALS FOR THORACIC MALIGNANCIES PHASE II PROOF-OF-CONCEPT STUDY OF PAZOPANIB MONOTHERAPY IN TREATMENT-NAIVE PATIENTS WITH STAGE I/II RESECTABLE NON–SMALL CELL LUNG CANCER Altorki N, Lane M, Bauer T, et al: J Clin Oncol 28:3131-3137, 2010 Altorki et al10 (Fig. 3) performed the next level of non–small cell lung cancer (NSCLC window-ofopportunity trials by using the antiangiogenic inhibitor pazopanib and the novel radiographic end point of high-resolution computed tomography (HRCT) scanning results to evaluate clinical responses to neoadjuvant pazopanib as the primary end point. Overall, 86% of patients had some amount of tumorvolume reduction after pazopanib therapy. This discovery trial incorporated translational studies into the study design with microarray gene expression analysis conducted on tumor specimens and blood specimens profiled for cytokine and angiogenic factors. Altorki et al10 conducted a multicenter phase II single-arm study of patients with resectable stage I-II NSCLC and treated them with pazopanib (800 mg/ d) for 2-6 weeks, which was followed by surgical resection. Pazopanib is an antiangiogenic tyrosine kinase inhibitor that targets vascular endothelial growth factor receptor (VEGFR), platelet-derived growth factor receptor (PDGFR), and the transmembrane proto-oncogene c-Kit. Patients underwent HRCT and had their blood collected before and after pazopanib therapy. The primary end point of the trial was the response rate as determined by HRCT (defined as Z50% tumor-volume reduction). Tumors were evaluated in the transaxial plane and the boundary drawn by a radiologist. The number of voxels comprising the image was summed to derive
the volume measurement. Secondary end points included response rate as determined by response evaluation criteria in solid tumors (RECIST), overall survival rate, and progression-free survival rate. Exploratory end points included gene expression analysis of the tumor tissue and profiling of plasma cytokine and angiogenic factors. A total of 35 patients with any NSCLC histology were enrolled, but only 86% completed the treatment of Z2 weeks of pazopanib. The median time of pazopanib use was 16 days. Among the patients, 4 patients discontinued pazopanib owing to adverse events, and 1 patient withdrew because of earlier surgical scheduling. None of the patients experienced surgical complications related to pazopanib's antiangiogenic activity. Clinical results revealed that 30 patients (86%) had some amount of tumor-volume reduction. Of the 35 patients, 2 (6%) had Z50% tumor-volume reduction and 23 (66%) had Z10% tumor-volume reduction. Moreover, 3 patients had a partial response as determined by RECIST. There was 1 patient who had discordant HRCT and RECIST criteria results: disease progression (Z20% increase in tumor diameter) but a 3% tumor-volume reduction. Tumor response did not correlate with the duration of neoadjuvant treatment. Of the 4 patients with squamous cell carcinoma, 3 had tumor-volume reduction. The 1 case with a complete pathologic response had a tumor-volume reduction of 86%. The 5 patients with extensive necrosis or fibrosis all experienced tumor-volume reduction as determined by HRCT. In the biomarker analysis, gene expression profiles showed increases in VEGFR-A (4.3-fold), PDGFR-α (10.3-fold), and PDGFR-β (4.1-fold), but none of these correlated with clinical response. In the
Figure 3. Phase II trial schema of neoadjuvant pazopanib for patients with resectable stage I-II NSCLC.10 FNA, fine needle aspiration; RR, response rate.
326
Seminars in Thoracic and Cardiovascular Surgery Volume 26, Number 4
WINDOW-OF-OPPORTUNITY TRIALS FOR THORACIC MALIGNANCIES exploratory serum analysis,11 blood samples were collected before and after neoadjuvant therapy. Baseline levels of 11 cytokine and angiogenic factors were associated with tumor shrinkage, with IL-12 having the strongest association. Posttreatment plasma-soluble VEGFR2 and IL-4 levels were also associated with tumor shrinkage. In the multivariate analysis, the levels of hepatocyte growth factor and IL-12 were associated with response with 81% accuracy. COMMENTARY The future of systemic therapy in NSCLC will depend on scientific discovery and the identification of biomarker targets and paired novel agents. The goal of personalized therapy is to identify the oncogenic genetic drivers in each patient and inhibit the biologic process with targeted agents or immunotherapy. Systemic chemotherapy algorithms will ultimately be supplanted by repeat biopsies of progressive disease and biomarker profiling, which will determine the next agent for treatment. To achieve this goal, window-of-opportunity trials are a necessary element of biomarker research. The 3 highlighted novel targeted agent window-ofopportunity trials described earlier are examples of the evolution of translational science in the field of thoracic oncology as biomarker assays are developed and become more capable of real-time use. Window-of-opportunity studies in NSCLC can be categorized as “discovery trials” or “validation trials,” Discovery trials are needed when the biomarker for the targeted agent is unknown, the treatment effect is uncertain, the study end points are unclear, and no molecular selection for patient enrollment is possible. These are by nature exploratory studies and should have strong translational components built into them to justify the risk to patients. It can be argued that only patients with early-stage (stage IA) NSCLC should be enrolled into these studies, as there is a higher risk of disease progression when the effect of the agent is unknown and there is no biomarker selection for enrollment. Validation trials are more mature: the biomarker-drug pairing is known, short-term treatment effects are more likely to be seen, and patient selection by a biomarker for enrollment is feasible for both neoadjuvant and adjuvant maintenance therapies. In validation trials, both confirmatory and exploratory translational studies are feasible. The study end points could be radiographic or biomarker related, or of both types. Also, patients with more advanced local-regional disease can be enrolled into these trials. The advantages of using targeted therapy windowof-opportunity trials to conduct clinical research are
significant. (1) These trials are designed to have a smaller sample size and can thus be completed more quickly. (2) There is an opportunity to obtain “pure” biomarker results; assessing the tumor for biomarkers before and after neoadjuvant treatment specific to the targeted agent allows for uncontaminated findings. This is also the case for clean assessment of peripheral surrogate biomarkers (eg, serum or platelets). (3) The radiographic response to the novel targeted agent under investigation can be directly correlated with radiographic and pathologic results with the least amount of confounding variables. (4) There is the potential to evaluate tumor heterogeneity and response to the targeted agent. However, there are major disadvantages to these trials as well. (1) These trials need to be conducted at experienced centers with multidisciplinary programs and strong thoracic surgery programs with established infrastructures for clinical research. (2) The small sample size needs a greater effect to be statistically significant. (3) Ideally, the biomarker target of the novel agent should be known to select patients and enhance response rates preoperatively. If the target is not known, there is a risk to the patients for disease progression and ineligibility for surgical resection. As mentioned earlier, discovery trials should likely be conducted with only patients with very early-stage NSCLC to limit this risk. (4) The novel agent must have a reasonable safety profile and tolerability, which requires prior human clinical trials to establish this toxicity profile and therefore limits the selection of agents to those which have already been developed in the setting of advanced stage disease. (5) A short time frame for neoadjuvant treatment is needed to prevent surgical delays. A small amount of neoadjuvant drug exposure is unlikely to affect survival outcomes and therefore these window-of-opportunity trials need to include adjuvant maintenance therapy to reveal any survival benefit. This situation poses additional challenges because it is still unknown what the optimal duration of adjuvant maintenance therapy should be with targeted agents for patients with early-stage NSCLC. (6) Alternative end points for these trials are needed, as the short duration of neoadjuvant therapy is unlikely to lead to significant response rates as determined by response evaluation criteria in solid tumors (RECIST). The only exception is when the biomarker-targeted agent pair has a similar rapid and large effect, as is the case of epidermal growth factor receptor (EGFR) mutations and EGFR tyrosine kinase inhibitors. Identifying alternative end points is a significant hurdle because the molecular target or novel radiographic imaging technique may not have been validated yet in prospective studies.
Seminars in Thoracic and Cardiovascular Surgery Volume 26, Number 4
327
WINDOW-OF-OPPORTUNITY TRIALS FOR THORACIC MALIGNANCIES Altorki et al10 designed their study to use tumorvolume reduction Z50% as determined from highresolution computed tomography scans for their end point. This end point, while innovative and capable of demonstrating minor changes in the anatomical size of tumors, conflicted with the results determined using RECIST in 1 patient, in which the tumor shape changed in response to pazopanib. Advances in radiographic or molecular measurements are clearly needed to complement window-of-opportunity trials. Several potential clinical end points can currently be used in window-of-opportunity trials. The most common end points are used in other trial designs as well and include survival outcomes and radiographic imaging results.12,13 Quality of life, cost or value of surgical resection, and days in the hospital are additional potential end points. Radiographic response, patientreported outcomes, and overall survival and progression-free survival rates are traditionally accepted end points for the U.S. Food and Drug Administration (FDA) drug approval.14 However, these traditional end points are less informative and more difficult to use in small clinical trials. Also, traditional RECIST responses are unlikely to be seen during the short duration of neoadjuvant therapy (2-4 weeks on average). Alternative ways to assess radiographic responses are therefore needed. Positron emission tomography-computed tomography (PET-CT) scan results and standard uptake value (SUV) levels have been proposed to measure neoadjuvant therapy response outcomes in patients with NSCLC, mesothelioma, or thymoma.15-17 However, the use of PET-CT results and SUV levels for this application is controversial owing to numerous confounding factors (eg, inflammation, postradiation effects, infection, or aspiration). Advances in software technology and reliable and consistent measurement techniques, specifically volumetric measurements, are needed to validate PET-CT as a dependable tool for end point analysis. Alternative end points are becoming more prominent in clinical trial designs and, although these end
points might not be approved by the FDA, accelerated pathways to approval are given when the biomarker (eg, tumor or blood) is known and patients who are likely to receive clinical benefit from a novel agent can be selected. Molecular end points using modulation of the tumor or a peripheral surrogate biomarker can be implemented in the trial design. Some examples of the technology that can be used to evaluate the molecular end points are included in Table 2. At The University of Texas MD Anderson Cancer Center, a neoadjuvant dasatinib window-of-opportunity trial for patients with resectable malignant pleural mesothelioma (MPM) was conducted in which the primary end point was modulation of p-SrcTyr419 in tumor cells. Patients who exhibited biomarker modulation or a response to 4 weeks of neoadjuvant dasatinib were eligible for maintenance dasatinib for 2 years. This study design was based on preclinical studies of mesothelioma cell lines showing that p-SrcTyr419 was overexpressed in MPM and that dasatinib had antitumor effects that corresponded to dephosphorylation of the biomarker p-SrcTyr419.18 The challenge with using molecular biomarkers as trial end points is that preclinical studies must confirm that the biomarker is valid and that the novel agent has target specificity to the tumor type under evaluation. An estimate of the amount of biomarker modulation owing to inhibition by the agent is also needed to determine appropriate statistical analyses for the trial. Knowing the target for the novel agent and being able to select patients likely to respond to therapy are crucial. Studies of pathologic end points other than tumor biomarkers have been reported. William et al19 reported that the percentage of residual viable tumor may be better at predicting survival than CT scan response is.19 In that study, 160 patients with resected NSCLC who had undergone neoadjuvant chemotherapy and then surgery were evaluated retrospectively. Preneoadjuvant and postneoadjuvant CT scans and the percentage of viable tumor in the pathologic specimens were compared with
Table 2. Potential Technologies for Analyzing Tumor or Peripheral Surrogate Biomarker End Points Technologies to Analyze Potential End Points CLIA standard assays (EGFR mutations, EML-4 ALK, and K-ras mutations) High-throughput mutational analysis (next-generation sequencing) mRNA and miRNA profiling SNP array for gene copy number Proteomics studies (RPPA and MS-MALDI) Plasma-serum analysis (cytokine and angiogenic factor profiling with multiplex bead arrays) Circulating tumor cell count (mutation analysis) CLIA, clinical laboratory improvement amendments; EML-4 ALK, echinoderm microtubule associated protein like 4anaplastic lymphoma kinase; MALDI, matrix-assisted laser desorption/ionization; mRNA, messenger RNA; miRNA, microRNA; MS, mass spectrometry; RPPA, reverse-phase protein array; SNP, single nucleotide polymorphism.
328
Seminars in Thoracic and Cardiovascular Surgery Volume 26, Number 4
WINDOW-OF-OPPORTUNITY TRIALS FOR THORACIC MALIGNANCIES survival outcomes. The multivariable analysis demonstrated that only pathologic stage and histopathologic response were significant predictors of overall survival. There was a 41% overall discordance rate between CT response as determined by RECIST and histopathologic response.
Future Neoadjuvant Thoracic Malignancy Trials Several notable neoadjuvant trials with NSCLC or mesothelioma patients are ongoing or under development. The ATTAIN (NCT01480141) trial (A windowof-opportunity trial of afatinib in early-stage NSCLC) is designed to administer afatinib for 14 days as neoadjuvant therapy before surgical resection for patients with stage IA-IIIA. Afatinib is an oral irreversible EGFR tyrosine kinase inhibitor that was recently approved by the FDA for front-line use in patients with metastatic NSCLC with mutated EGFR. Patients will receive a core tumor biopsy at baseline followed by neoadjuvant afatinib and then undergo surgical resection. PET-CT scans will be scheduled at baseline and preoperatively. The primary end point is feasibility and completion of 14 days of afatinib treatment and then thoracotomy. Secondary end points include PET-CT response, SUV radiographic response, safety, and efficacy. Translational studies of preneoadjuvant and postneoadjuvant afatinib tumor specimens are exploratory. A phase I-II trial comprising patients with NSCLC with stage IB-IIIA disease will be studying the effect of nintedanib, an oral vascular endothelial growth factor receptor, platelet-derived growth factor receptor, and fibroblast growth factor receptor angiokinase inhibitor. The phase I portion has a nintedanib monotherapy run-in followed by 3 cycles of chemotherapy plus nintedanib before surgical resection. Baseline biopsies and radiographic imaging are included in the study. The phase II portion will be conducted through the Southwest Oncology Group and will enroll 140 patients with resectable NSCLC (stage IB [Z4 cm]IIIA). Patients will be randomly allocated to treatment with cisplatin-docetaxel with or without nintedanib for 3 cycles of therapy followed by surgical resection. The primary end point to the phase II portion is the major pathologic response. Secondary end points will include safety-tolerability, radiographic response, and survival outcomes. Finally, a neoadjuvant Southwest Oncology Group study is under development, which will enroll patients with resectable epithelioid or biphasic MPM and randomly assign them to cisplatin-pemetrexed with or without MPDL3280A, a PD-L1 inhibitor. Patients in the MPDL3280A arm will receive 1 year of
maintenance therapy after resection and adjuvant radiation therapy. The primary end point of this study is progression-free survival, and secondary end points included response (according to immune RECIST and modified RECIST) and quality of life after tumor resection. The translational studies will include programmed death 1 (PD-1)/PD-L1 immunohistochemistry (IHC) expression in tumor cells, serum cytokine analysis, and gene expression profiling of plasma. PD-1 protein, a T-cell coinhibitory receptor, and one of its ligands, PD-L1, are immunotherapy targets. The PD-1 receptor binds PD-L1 and prevents T-cell inhibition and downregulates T-cell responses. Several solid tumor studies have shown that when the PD-1/PD-L1 interaction is blocked, T-cell activity is restored with a corresponding antitumor effect.20 Several PD-L1 inhibitors are under development, including BMS0936559 (Bristol-Meyers Squibb), MEDI-4736 (Medimmune), and MPDL3280A (Genentech).21-23 In NSCLC studies using these PD-L1 inhibitors as monotherapy in phase I-II trials, response rates have ranged from 10%26%.20-23 Although there are no mesothelioma studies to date, MPM is a highly immunogenic disease, so treatment may benefit from the addition of an immunotherapy agent to neoadjuvant therapy and, as maintenance treatment, may activate T cells against microscopic residual disease.24 Mansfield et al25 reported in a retrospective archival MPM tumor tissue analysis that MPM (n ¼ 224) had 40% PD-L1 IHC expression as determined using mouse monoclonal anti–human B7-H1 antibody (clone 5H1-A3). Moreover, PD-L1 IHC expression was associated with greater disease burden and a worse survival outcome (6 vs 14 months, P o 0.0001).25 CONCLUSION In conclusion, window-of-opportunity studies with targeted agents have an important role in advancing the study of early-stage NSCLC and mesothelioma. There is great potential for direct translational correlations because these studies provide “pure” results and expeditious outcomes owing to small sample sizes. However, these window-ofopportunity trials may not necessarily benefit the patient unless the patient has an actionable mutation and the magnitude of benefit derived from the novel agent is high. Thus, larger and more generalized neoadjuvant trials are also needed with welldesigned correlative science. These trials are more likely to provide patients without actionable mutations with clinical benefit and could further our understanding of the biology of this disease, albeit with confounding factors. Both trial concepts require the support and expertise of thoracic surgery to ensure optimal clinical and translational results.
Seminars in Thoracic and Cardiovascular Surgery Volume 26, Number 4
329
WINDOW-OF-OPPORTUNITY TRIALS FOR THORACIC MALIGNANCIES 1. NSCLC Meta-Analysis Collaborative Group: Preoperative chemotherapy for non-small-cell lung cancer: A systematic review and metaanalysis of individual participant data. Lancet 383:1561-1571, 2014 2. Albain KS, Swann RS, Rusch VW, et al: Radiotherapy plus chemotherapy with or without surgical resection for stage III non-small-cell lung cancer: A phase III randomised controlled trial. Lancet 374:379-386, 2009 3. Betticher DC, Hsu Schmitz SF, Totsch M, et al: Mediastinal lymph node clearance after docetaxel-cisplatin neoadjuvant chemotherapy is prognostic of survival in patients with stage IIIA pN2 non-small-cell lung cancer: A multicenter phase II trial. J Clin Oncol 21: 1752-1759, 2003 4. Betticher DC, Hsu Schmitz SF, Totsch M, et al: Prognostic factors affecting long-term outcomes in patients with resected stage IIIA pN2 non-small-cell lung cancer: 5-Year follow-up of a phase II study. Br J Cancer 94: 1099-1106, 2006 5. Donington JS, Pass HI: Surgical approach to locally advanced non-small cell lung cancer. Cancer J 19:217-221, 2013 6. Lara-Guerra H, Waddell TK, Salvarrey MA, et al: Phase II study of preoperative gefitinib in clinical stage I non-small-cell lung cancer. J Clin Oncol 27:6229-6236, 2009 7. Rizvi NA, Rusch V, Pao W, et al: Molecular characteristics predict clinical outcomes: Prospective trial correlating response to the EGFR tyrosine kinase inhibitor gefitinib with the presence of sensitizing mutations in the tyrosine binding domain of the EGFR gene. Clin Cancer Res 17:3500-3506, 2011 8. Lynch TJ, Bell DW, Sordella R, et al: Activating mutations in the epidermal growth factor receptor underlying responsiveness of non–
330
9.
10.
11.
12.
13.
14.
15.
16.
small-cell lung cancer to gefitinib. N Engl J Med 350:2129-2139, 2004 Paez JG, Janne PA, Lee JC, et al: EGFR mutations in lung cancer: Correlation with clinical response to gefitinib therapy. Science 304:1497-1500, 2004 Altorki N, Lane ME, Bauer T, et al: Phase II proof-of-concept study of pazopanib monotherapy in treatment-naive patients with stage I/II resectable non-small-cell lung cancer. J Clin Oncol 28:3131-3137, 2010 Nikolinakos PG, Altorki N, Yankelevitz D, et al: Plasma cytokine and angiogenic factor profiling identifies markers associated with tumor shrinkage in early-stage non-small cell lung cancer patients treated with pazopanib. Cancer Res 70:2171-2179, 2010 Machtay M, Paulus R, Moughan J, et al: Defining local-regional control and its importance in locally advanced non-small cell lung carcinoma. J Thorac Oncol 7:716-722, 2012 Mauguen A, Pignon JP, Burdett S, et al: Surrogate endpoints for overall survival in chemotherapy and radiotherapy trials in operable and locally advanced lung cancer: A re-analysis of meta-analyses of individual patients' data. Lancet Oncol 14:619-626, 2013 Pazdur R: Endpoints for assessing drug activity in clinical trials. Oncologist 13:19-21, 2008 (suppl 2) Benveniste MF, Moran CA, Mawlawi O, et al: FDG PET-CT aids in the preoperative assessment of patients with newly diagnosed thymic epithelial malignancies. J Thorac Oncol 8: 502-510, 2013 Tsutani Y, Takuwa T, Miyata Y, et al: Prognostic significance of metabolic response by positron emission tomography after neoadjuvant chemotherapy for resectable malignant pleural mesothelioma. Ann Oncol 24:1005-1010, 2013
17. Lee HY, Lee KS, Park J, et al: Baseline SUVmax at PET-CT in stage IIIA non-small-cell lung cancer patients undergoing surgery after neoadjuvant therapy: Prognostic implication focused on histopathologic subtypes. Acad Radiol 19:440-445, 2012 18. Tsao AS, He D, Saigal B, et al: Inhibition of c-Src expression and activation in malignant pleural mesothelioma tissues leads to apoptosis, cell cycle arrest, and decreased migration and invasion. Mol Cancer Ther 6:1962-1972, 2007 19. William Jr. WN, Pataer A, Kalhor N, et al: Computed tomography RECIST assessment of histopathologic response and prediction of survival in patients with resectable non-smallcell lung cancer after neoadjuvant chemotherapy. J Thorac Oncol 8:222-228, 2013 20. Brahmer JR: Harnessing the immune system for the treatment of non-small-cell lung cancer. J Clin Oncol 31:1021-1028, 2013 21. Brahmer JR, Tykodi SS, Chow LQ, et al: Safety and activity of anti-PD-L1 antibody in patients with advanced cancer. N Engl J Med 366: 2455-2465, 2012 22. Brahmer J, Rizvi N, Lutzky J, et al: Clinical activity and biomarkers of MEDI4736, an antiPD-L1 antibody, in patients with NSCLC. J Clin Oncol ;32 [Abstract 8021] 23. Davies M: New modalities of cancer treatment for NSCLC: Focus on immunotherapy. Cancer Manage Res 6:63-75, 2014 24. Thomas A, Hassan R: Immunotherapies for non-small-cell lung cancer and mesothelioma. Lancet Oncol 13:e301-e310, 2012 25. Mansfield A, Peikert T, Roden A, et al: Programmed cell death 1 ligand 1 expression and association with survival in mesothelioma. J Thorac Oncol 9:S7-S52, 2014
Seminars in Thoracic and Cardiovascular Surgery Volume 26, Number 4
Current Readings: Window-of-Opportunity Trials for Thoracic Malignancies Anne S. Tsao, MD Recent major advances in metastatic non–small cell lung cancer have occurred with the identification of molecular biomarker targets and administration of novel agents with resulting improvement in clinical outcomes. In the early-stage setting, personalized therapy with novel agents and molecular profiling are being incorporated into neoadjuvant “window-of-opportunity” trials. These important studies enable biomarker research and an expedited analysis of the efficacy of the targeted agent. However, there are significant limitations to window-of-opportunity trials. The aim of this article is to review the current window-of-opportunity trials of neoadjuvant targeted agents for thoracic malignancies, discuss the benefits and limitations of these trials, and propose more optimal alternative trial end points. Neoadjuvant trials of resectable non–small cell lung cancer and mesothelioma that are ongoing or under development and relevant to thoracic surgeons are also discussed. The success of these trials will depend on a collaborative multidisciplinary effort, especially from the field of thoracic surgery. Semin Thoracic Surg 26:323–330 I 2014 Elsevier Inc. All rights reserved. Keywords: NSCLC, window-of-opportunity trials, targeted agents, neoadjuvant, thoracic malignancies INTRODUCTION Neoadjuvant chemotherapy is a common approach to managing resectable local or regionally advanced non–small cell lung cancer (NSCLC). Most neoadjuvant trials have used 4 cycles of platinum-doublet chemotherapy and, as a whole, have shown a 5% absolute survival benefit at 5 years.1 The advantages and disadvantages to using neoadjuvant chemotherapy are listed in Table 1. Clinical trials using neoadjuvant chemotherapy or chemoradiotherapy have identified prognostic factors for improved survival after tumor resection.2-5 Patients with superior survival outcomes include those who achieve a response to chemotherapy, whose disease is downstaged from N2, who have single-station N2 rather than multistation N2 disease, who achieve R0 resection, and who have a complete pathologic response.2,4 Department of Thoracic & Head and Neck Medical Oncology, The University of Texas MD Anderson Cancer Center, Houston, Texas Dr Tsao reports receiving consulting fees from, Roche, Genentech, Novartis, Boehringer-Ingelheim, Medimmune, and Lilly. Address reprint requests to Anne S. Tsao, MD, Department of Thoracic & Head and Neck Medical Oncology, The University of Texas MD Anderson Cancer Center, 1515 Holcombe Blvd., Unit 432, Houston, TX 77030. E-mail: [email protected]
1043-0679/$-see front matter ª 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1053/j.semtcvs.2014.12.005
Thus, it can be extrapolated that better systemic therapies may lead to increased downstaging and superior clinical outcomes. In the era of personalized therapy with novel agents and molecular profiling, designing studies—such as “window-of-opportunity” trials—that enable biomarker research and expedited analysis of whether the agents are efficacious are critically needed. This review article summarizes the current literature on 3 thoracic window-of-opportunity studies with novel targeted agents to illustrate the evolution of this type of trial design. The advantages and disadvantages of novel agent window-ofopportunity trials and alternative trial end points are discussed. Ongoing neoadjuvant studies with novel agents relevant to the thoracic surgical community are summarized. Window-of-opportunity studies with multidisciplinary effort, especially surgical support, are a necessary and vital component to move the field of thoracic oncology forward. PHASE II STUDY OF PREOPERATIVE GEFITINIB IN CLINICAL STAGE I NON– SMALL CELL LUNG CANCER Lara-Guerra H, Waddell T, Salvarrey M, et al: J Clin Oncol 27:6229-6236, 2009 Lara-Guerra et al6 conducted a preoperative window-of-opportunity trial (Fig. 1) using gefitinib, an epidermal growth factor receptor (EGFR) tyrosine 323
WINDOW-OF-OPPORTUNITY TRIALS FOR THORACIC MALIGNANCIES Table 1. Hypothetical Implications of Neoadjuvant Therapy Advantages
Disadvantages
Gives patients time to quit using tobacco Eliminates micrometastatic disease earlier
Delays definitive procedure Interferes with surgery owing to the toxicity of chemotherapy Creates staging ambiguity
Allows higher dose chemotherapy (patients tolerate chemotherapy better before rather than after major surgery) May lead to downstaging of disease Provides prognostic value (assessment of chemosensitivity)
kinase inhibitor, and demonstrated that this type of trial design is feasible and safe for patients with earlystage resectable non–small cell lung cancer (NSCLC). In this single-arm phase II study, Lara-Guerra et al6 enrolled 36 patients with stage I resectable NSCLC and treated them with preoperative gefitinib (250 mg/d) for up to 28 days. Radiographic imaging results as determined using standard response evaluation criteria in solid tumors (RECIST) were obtained before and after preoperative gefitinib therapy. After 24-48 hours from the last dose of neoadjuvant gefitinib, patients underwent mediastinoscopy and surgical resection. The primary end point of the trial was overall response rate as determined by RECIST, and the secondary end points included the overall survival and progression-free survival rates. The translational studies included analysis of tumor tissue for EGFR by immunohistochemistry and fluorescent in situ hybridization, EGFR gene mutation, and serum levels for tumor growth factor α. Among the 35 evaluable patients, 11% had a partial response, 81% stable disease, and 8% progression of disease as determined by RECIST after 28
Increases the risk of postoperative complications
days of gefitinib treatment. Tumors shrank in 15 (43%) patients, grew in 15 patients (43%), and did not change in size in 5 (14%) patients. The only biomarker that predicted a clinical partial response was the presence of a sensitizing EGFR mutation (deletion exon 19 or L858 mutation). Among the patients with a KRAS mutation, there were no responders. No other biomarkers were predictive or prognostic, although in one case a patient with a high EGFR gene copy number as determined by fluorescent in situ hybridization and no EGFR mutation also demonstrated a partial response.
MOLECULAR CHARACTERISTICS PREDICT CLINICAL OUTCOMES: PROSPECTIVE TRIAL CORRELATING RESPONSE TO THE EPIDERMAL GROWTH FACTOR RECEPTOR TYROSINE KINASE INHIBITOR GEFITINIB WITH THE PRESENCE OF SENSITIZING MUTATIONS IN THE TYROSINE BINDING DOMAIN OF THE EGFR GENE Rizvi N, Rusch V, Pao W, et al: Clin Cancer Res 17:3500-3506, 2011
Figure 1. Phase II trial schema of preoperative gefitinib for patients with stage I resectable NSCLC.6 FISH, fluorescent in situ hybridization; FNA, fine needle aspiration; ORR, overall response rate; OS, overall survival; PFS, progression-free survival; TGFα, transforming growth factor α.
324
Seminars in Thoracic and Cardiovascular Surgery Volume 26, Number 4
WINDOW-OF-OPPORTUNITY TRIALS FOR THORACIC MALIGNANCIES Rizvi et al7 (Fig. 2) took non–small cell lung cancer (NSCLC) window-of-opportunity trials to the next step in this study that selected patients by clinical characteristics to enhance the likelihood of enrolling patients who had epidermal growth factor receptor (EGFR) mutations. This study verified that the presence of a sensitizing EGFR mutation correlated with a radiographic response. However, there was no difference in survival outcomes, which was likely owing to the short duration of neoadjuvant gefitinib therapy and the inability of many patients to tolerate adjuvant maintenance gefitinib therapy for 2 years. Rizvi et al7 enrolled patients with stage I-II NSCLC who had a o15 pack-year smoking history, a bronchioloalveolar component on histologic assessment, or both to a single-arm trial. The clinical selection was designed to enrich the population with sensitizing EGFR mutations, which at the time of the trial design had just been identified to be predictive for dramatic responses to EGFR tyrosine kinase inhibitors.8,9 In this trial, patients with NSCLC were treated with gefitinib therapy (250 mg/d) for up to 21 days and 48 hours later underwent surgical resection and adjuvant chemotherapy if appropriate. Radiographic images were obtained before and after gefitinib therapy, and the tumors were measured using the World Health Organization (WHO) bidimensional measurement criteria. Patients who had a radiographic response or identification of a sensitizing EGFR mutation were offered adjuvant maintenance gefitinib (250 mg/d) for up to 2 years. Patients who did not have an EGFR mutation and did not have a radiographic response to gefitinib were observed after resection. The primary end point was to correlate the WHO bidimensional radiographic response to EGFR mutation status. Secondary end points included overall survival and progression-free survival rates.
A total of 50 patients were enrolled (39 women and 11 men). Among them, 36% of patients were never smokers and 46% were r15 pack-year former smokers. There were 9 patients who had bronchioloalveolar features on histologic analysis. In total, 21 patients had a radiographic response to neoadjuvant gefitinib, all of whom had undergone surgical resection, whereas among the 29 non-responders, 3 could not undergo resection owing to upstaging. Of the 47 patients who underwent surgery, 6 also received adjuvant chemotherapy. Postoperatively, adjuvant gefitinib maintenance therapy was offered to the 21 responding patients, which was initiated on postoperative day 7. Postoperative complications included 2 cases of pneumonitis, 2 of atrial fibrillation, 1 of supraventricular arrhythmia, 3 of air leaks, and 1 of intraoperative injury to the right main pulmonary artery. The clinical results included 21 patients with Z25% tumor shrinkage by WHO bidimensional measurements after 21 days of gefitinib therapy. Of the 21 responding patients, 17 had an EGFR mutation. In addition, 5 patients had a KRAS mutation, and none had a radiographic response. There was no correlation between EGFR immunohistochemistry results or gene amplification. In one curious case, the tumor of a patient with a PI3KCA (E542K) mutation and a BRAF (V600E) mutation shrank by 74% in response to gefitinib therapy. In this trial, the median progression-free survival rate was not reached. The analysis was compromised because less than half of the 21 patients eligible for the 2 years of adjuvant gefitinib were able to complete the therapy. The 2-year disease-free survival rate was 95% for patients who received adjuvant gefitinib and 78% for patients who did not. The 2-year disease-free survival rate was 95% for patients with an EGFR mutation and 90% for those who did not.
Figure 2. Trial schema of preoperative gefitinib for EGFR mutation–enriched population of patients with stage I-II NSCLC.7 BAC, bronchioloalveolar carcinoma.
Seminars in Thoracic and Cardiovascular Surgery Volume 26, Number 4
325
WINDOW-OF-OPPORTUNITY TRIALS FOR THORACIC MALIGNANCIES PHASE II PROOF-OF-CONCEPT STUDY OF PAZOPANIB MONOTHERAPY IN TREATMENT-NAIVE PATIENTS WITH STAGE I/II RESECTABLE NON–SMALL CELL LUNG CANCER Altorki N, Lane M, Bauer T, et al: J Clin Oncol 28:3131-3137, 2010 Altorki et al10 (Fig. 3) performed the next level of non–small cell lung cancer (NSCLC window-ofopportunity trials by using the antiangiogenic inhibitor pazopanib and the novel radiographic end point of high-resolution computed tomography (HRCT) scanning results to evaluate clinical responses to neoadjuvant pazopanib as the primary end point. Overall, 86% of patients had some amount of tumorvolume reduction after pazopanib therapy. This discovery trial incorporated translational studies into the study design with microarray gene expression analysis conducted on tumor specimens and blood specimens profiled for cytokine and angiogenic factors. Altorki et al10 conducted a multicenter phase II single-arm study of patients with resectable stage I-II NSCLC and treated them with pazopanib (800 mg/ d) for 2-6 weeks, which was followed by surgical resection. Pazopanib is an antiangiogenic tyrosine kinase inhibitor that targets vascular endothelial growth factor receptor (VEGFR), platelet-derived growth factor receptor (PDGFR), and the transmembrane proto-oncogene c-Kit. Patients underwent HRCT and had their blood collected before and after pazopanib therapy. The primary end point of the trial was the response rate as determined by HRCT (defined as Z50% tumor-volume reduction). Tumors were evaluated in the transaxial plane and the boundary drawn by a radiologist. The number of voxels comprising the image was summed to derive
the volume measurement. Secondary end points included response rate as determined by response evaluation criteria in solid tumors (RECIST), overall survival rate, and progression-free survival rate. Exploratory end points included gene expression analysis of the tumor tissue and profiling of plasma cytokine and angiogenic factors. A total of 35 patients with any NSCLC histology were enrolled, but only 86% completed the treatment of Z2 weeks of pazopanib. The median time of pazopanib use was 16 days. Among the patients, 4 patients discontinued pazopanib owing to adverse events, and 1 patient withdrew because of earlier surgical scheduling. None of the patients experienced surgical complications related to pazopanib's antiangiogenic activity. Clinical results revealed that 30 patients (86%) had some amount of tumor-volume reduction. Of the 35 patients, 2 (6%) had Z50% tumor-volume reduction and 23 (66%) had Z10% tumor-volume reduction. Moreover, 3 patients had a partial response as determined by RECIST. There was 1 patient who had discordant HRCT and RECIST criteria results: disease progression (Z20% increase in tumor diameter) but a 3% tumor-volume reduction. Tumor response did not correlate with the duration of neoadjuvant treatment. Of the 4 patients with squamous cell carcinoma, 3 had tumor-volume reduction. The 1 case with a complete pathologic response had a tumor-volume reduction of 86%. The 5 patients with extensive necrosis or fibrosis all experienced tumor-volume reduction as determined by HRCT. In the biomarker analysis, gene expression profiles showed increases in VEGFR-A (4.3-fold), PDGFR-α (10.3-fold), and PDGFR-β (4.1-fold), but none of these correlated with clinical response. In the
Figure 3. Phase II trial schema of neoadjuvant pazopanib for patients with resectable stage I-II NSCLC.10 FNA, fine needle aspiration; RR, response rate.
326
Seminars in Thoracic and Cardiovascular Surgery Volume 26, Number 4
WINDOW-OF-OPPORTUNITY TRIALS FOR THORACIC MALIGNANCIES exploratory serum analysis,11 blood samples were collected before and after neoadjuvant therapy. Baseline levels of 11 cytokine and angiogenic factors were associated with tumor shrinkage, with IL-12 having the strongest association. Posttreatment plasma-soluble VEGFR2 and IL-4 levels were also associated with tumor shrinkage. In the multivariate analysis, the levels of hepatocyte growth factor and IL-12 were associated with response with 81% accuracy. COMMENTARY The future of systemic therapy in NSCLC will depend on scientific discovery and the identification of biomarker targets and paired novel agents. The goal of personalized therapy is to identify the oncogenic genetic drivers in each patient and inhibit the biologic process with targeted agents or immunotherapy. Systemic chemotherapy algorithms will ultimately be supplanted by repeat biopsies of progressive disease and biomarker profiling, which will determine the next agent for treatment. To achieve this goal, window-of-opportunity trials are a necessary element of biomarker research. The 3 highlighted novel targeted agent window-ofopportunity trials described earlier are examples of the evolution of translational science in the field of thoracic oncology as biomarker assays are developed and become more capable of real-time use. Window-of-opportunity studies in NSCLC can be categorized as “discovery trials” or “validation trials,” Discovery trials are needed when the biomarker for the targeted agent is unknown, the treatment effect is uncertain, the study end points are unclear, and no molecular selection for patient enrollment is possible. These are by nature exploratory studies and should have strong translational components built into them to justify the risk to patients. It can be argued that only patients with early-stage (stage IA) NSCLC should be enrolled into these studies, as there is a higher risk of disease progression when the effect of the agent is unknown and there is no biomarker selection for enrollment. Validation trials are more mature: the biomarker-drug pairing is known, short-term treatment effects are more likely to be seen, and patient selection by a biomarker for enrollment is feasible for both neoadjuvant and adjuvant maintenance therapies. In validation trials, both confirmatory and exploratory translational studies are feasible. The study end points could be radiographic or biomarker related, or of both types. Also, patients with more advanced local-regional disease can be enrolled into these trials. The advantages of using targeted therapy windowof-opportunity trials to conduct clinical research are
significant. (1) These trials are designed to have a smaller sample size and can thus be completed more quickly. (2) There is an opportunity to obtain “pure” biomarker results; assessing the tumor for biomarkers before and after neoadjuvant treatment specific to the targeted agent allows for uncontaminated findings. This is also the case for clean assessment of peripheral surrogate biomarkers (eg, serum or platelets). (3) The radiographic response to the novel targeted agent under investigation can be directly correlated with radiographic and pathologic results with the least amount of confounding variables. (4) There is the potential to evaluate tumor heterogeneity and response to the targeted agent. However, there are major disadvantages to these trials as well. (1) These trials need to be conducted at experienced centers with multidisciplinary programs and strong thoracic surgery programs with established infrastructures for clinical research. (2) The small sample size needs a greater effect to be statistically significant. (3) Ideally, the biomarker target of the novel agent should be known to select patients and enhance response rates preoperatively. If the target is not known, there is a risk to the patients for disease progression and ineligibility for surgical resection. As mentioned earlier, discovery trials should likely be conducted with only patients with very early-stage NSCLC to limit this risk. (4) The novel agent must have a reasonable safety profile and tolerability, which requires prior human clinical trials to establish this toxicity profile and therefore limits the selection of agents to those which have already been developed in the setting of advanced stage disease. (5) A short time frame for neoadjuvant treatment is needed to prevent surgical delays. A small amount of neoadjuvant drug exposure is unlikely to affect survival outcomes and therefore these window-of-opportunity trials need to include adjuvant maintenance therapy to reveal any survival benefit. This situation poses additional challenges because it is still unknown what the optimal duration of adjuvant maintenance therapy should be with targeted agents for patients with early-stage NSCLC. (6) Alternative end points for these trials are needed, as the short duration of neoadjuvant therapy is unlikely to lead to significant response rates as determined by response evaluation criteria in solid tumors (RECIST). The only exception is when the biomarker-targeted agent pair has a similar rapid and large effect, as is the case of epidermal growth factor receptor (EGFR) mutations and EGFR tyrosine kinase inhibitors. Identifying alternative end points is a significant hurdle because the molecular target or novel radiographic imaging technique may not have been validated yet in prospective studies.
Seminars in Thoracic and Cardiovascular Surgery Volume 26, Number 4
327
WINDOW-OF-OPPORTUNITY TRIALS FOR THORACIC MALIGNANCIES Altorki et al10 designed their study to use tumorvolume reduction Z50% as determined from highresolution computed tomography scans for their end point. This end point, while innovative and capable of demonstrating minor changes in the anatomical size of tumors, conflicted with the results determined using RECIST in 1 patient, in which the tumor shape changed in response to pazopanib. Advances in radiographic or molecular measurements are clearly needed to complement window-of-opportunity trials. Several potential clinical end points can currently be used in window-of-opportunity trials. The most common end points are used in other trial designs as well and include survival outcomes and radiographic imaging results.12,13 Quality of life, cost or value of surgical resection, and days in the hospital are additional potential end points. Radiographic response, patientreported outcomes, and overall survival and progression-free survival rates are traditionally accepted end points for the U.S. Food and Drug Administration (FDA) drug approval.14 However, these traditional end points are less informative and more difficult to use in small clinical trials. Also, traditional RECIST responses are unlikely to be seen during the short duration of neoadjuvant therapy (2-4 weeks on average). Alternative ways to assess radiographic responses are therefore needed. Positron emission tomography-computed tomography (PET-CT) scan results and standard uptake value (SUV) levels have been proposed to measure neoadjuvant therapy response outcomes in patients with NSCLC, mesothelioma, or thymoma.15-17 However, the use of PET-CT results and SUV levels for this application is controversial owing to numerous confounding factors (eg, inflammation, postradiation effects, infection, or aspiration). Advances in software technology and reliable and consistent measurement techniques, specifically volumetric measurements, are needed to validate PET-CT as a dependable tool for end point analysis. Alternative end points are becoming more prominent in clinical trial designs and, although these end
points might not be approved by the FDA, accelerated pathways to approval are given when the biomarker (eg, tumor or blood) is known and patients who are likely to receive clinical benefit from a novel agent can be selected. Molecular end points using modulation of the tumor or a peripheral surrogate biomarker can be implemented in the trial design. Some examples of the technology that can be used to evaluate the molecular end points are included in Table 2. At The University of Texas MD Anderson Cancer Center, a neoadjuvant dasatinib window-of-opportunity trial for patients with resectable malignant pleural mesothelioma (MPM) was conducted in which the primary end point was modulation of p-SrcTyr419 in tumor cells. Patients who exhibited biomarker modulation or a response to 4 weeks of neoadjuvant dasatinib were eligible for maintenance dasatinib for 2 years. This study design was based on preclinical studies of mesothelioma cell lines showing that p-SrcTyr419 was overexpressed in MPM and that dasatinib had antitumor effects that corresponded to dephosphorylation of the biomarker p-SrcTyr419.18 The challenge with using molecular biomarkers as trial end points is that preclinical studies must confirm that the biomarker is valid and that the novel agent has target specificity to the tumor type under evaluation. An estimate of the amount of biomarker modulation owing to inhibition by the agent is also needed to determine appropriate statistical analyses for the trial. Knowing the target for the novel agent and being able to select patients likely to respond to therapy are crucial. Studies of pathologic end points other than tumor biomarkers have been reported. William et al19 reported that the percentage of residual viable tumor may be better at predicting survival than CT scan response is.19 In that study, 160 patients with resected NSCLC who had undergone neoadjuvant chemotherapy and then surgery were evaluated retrospectively. Preneoadjuvant and postneoadjuvant CT scans and the percentage of viable tumor in the pathologic specimens were compared with
Table 2. Potential Technologies for Analyzing Tumor or Peripheral Surrogate Biomarker End Points Technologies to Analyze Potential End Points CLIA standard assays (EGFR mutations, EML-4 ALK, and K-ras mutations) High-throughput mutational analysis (next-generation sequencing) mRNA and miRNA profiling SNP array for gene copy number Proteomics studies (RPPA and MS-MALDI) Plasma-serum analysis (cytokine and angiogenic factor profiling with multiplex bead arrays) Circulating tumor cell count (mutation analysis) CLIA, clinical laboratory improvement amendments; EML-4 ALK, echinoderm microtubule associated protein like 4anaplastic lymphoma kinase; MALDI, matrix-assisted laser desorption/ionization; mRNA, messenger RNA; miRNA, microRNA; MS, mass spectrometry; RPPA, reverse-phase protein array; SNP, single nucleotide polymorphism.
328
Seminars in Thoracic and Cardiovascular Surgery Volume 26, Number 4
WINDOW-OF-OPPORTUNITY TRIALS FOR THORACIC MALIGNANCIES survival outcomes. The multivariable analysis demonstrated that only pathologic stage and histopathologic response were significant predictors of overall survival. There was a 41% overall discordance rate between CT response as determined by RECIST and histopathologic response.
Future Neoadjuvant Thoracic Malignancy Trials Several notable neoadjuvant trials with NSCLC or mesothelioma patients are ongoing or under development. The ATTAIN (NCT01480141) trial (A windowof-opportunity trial of afatinib in early-stage NSCLC) is designed to administer afatinib for 14 days as neoadjuvant therapy before surgical resection for patients with stage IA-IIIA. Afatinib is an oral irreversible EGFR tyrosine kinase inhibitor that was recently approved by the FDA for front-line use in patients with metastatic NSCLC with mutated EGFR. Patients will receive a core tumor biopsy at baseline followed by neoadjuvant afatinib and then undergo surgical resection. PET-CT scans will be scheduled at baseline and preoperatively. The primary end point is feasibility and completion of 14 days of afatinib treatment and then thoracotomy. Secondary end points include PET-CT response, SUV radiographic response, safety, and efficacy. Translational studies of preneoadjuvant and postneoadjuvant afatinib tumor specimens are exploratory. A phase I-II trial comprising patients with NSCLC with stage IB-IIIA disease will be studying the effect of nintedanib, an oral vascular endothelial growth factor receptor, platelet-derived growth factor receptor, and fibroblast growth factor receptor angiokinase inhibitor. The phase I portion has a nintedanib monotherapy run-in followed by 3 cycles of chemotherapy plus nintedanib before surgical resection. Baseline biopsies and radiographic imaging are included in the study. The phase II portion will be conducted through the Southwest Oncology Group and will enroll 140 patients with resectable NSCLC (stage IB [Z4 cm]IIIA). Patients will be randomly allocated to treatment with cisplatin-docetaxel with or without nintedanib for 3 cycles of therapy followed by surgical resection. The primary end point to the phase II portion is the major pathologic response. Secondary end points will include safety-tolerability, radiographic response, and survival outcomes. Finally, a neoadjuvant Southwest Oncology Group study is under development, which will enroll patients with resectable epithelioid or biphasic MPM and randomly assign them to cisplatin-pemetrexed with or without MPDL3280A, a PD-L1 inhibitor. Patients in the MPDL3280A arm will receive 1 year of
maintenance therapy after resection and adjuvant radiation therapy. The primary end point of this study is progression-free survival, and secondary end points included response (according to immune RECIST and modified RECIST) and quality of life after tumor resection. The translational studies will include programmed death 1 (PD-1)/PD-L1 immunohistochemistry (IHC) expression in tumor cells, serum cytokine analysis, and gene expression profiling of plasma. PD-1 protein, a T-cell coinhibitory receptor, and one of its ligands, PD-L1, are immunotherapy targets. The PD-1 receptor binds PD-L1 and prevents T-cell inhibition and downregulates T-cell responses. Several solid tumor studies have shown that when the PD-1/PD-L1 interaction is blocked, T-cell activity is restored with a corresponding antitumor effect.20 Several PD-L1 inhibitors are under development, including BMS0936559 (Bristol-Meyers Squibb), MEDI-4736 (Medimmune), and MPDL3280A (Genentech).21-23 In NSCLC studies using these PD-L1 inhibitors as monotherapy in phase I-II trials, response rates have ranged from 10%26%.20-23 Although there are no mesothelioma studies to date, MPM is a highly immunogenic disease, so treatment may benefit from the addition of an immunotherapy agent to neoadjuvant therapy and, as maintenance treatment, may activate T cells against microscopic residual disease.24 Mansfield et al25 reported in a retrospective archival MPM tumor tissue analysis that MPM (n ¼ 224) had 40% PD-L1 IHC expression as determined using mouse monoclonal anti–human B7-H1 antibody (clone 5H1-A3). Moreover, PD-L1 IHC expression was associated with greater disease burden and a worse survival outcome (6 vs 14 months, P o 0.0001).25 CONCLUSION In conclusion, window-of-opportunity studies with targeted agents have an important role in advancing the study of early-stage NSCLC and mesothelioma. There is great potential for direct translational correlations because these studies provide “pure” results and expeditious outcomes owing to small sample sizes. However, these window-ofopportunity trials may not necessarily benefit the patient unless the patient has an actionable mutation and the magnitude of benefit derived from the novel agent is high. Thus, larger and more generalized neoadjuvant trials are also needed with welldesigned correlative science. These trials are more likely to provide patients without actionable mutations with clinical benefit and could further our understanding of the biology of this disease, albeit with confounding factors. Both trial concepts require the support and expertise of thoracic surgery to ensure optimal clinical and translational results.
Seminars in Thoracic and Cardiovascular Surgery Volume 26, Number 4
329
WINDOW-OF-OPPORTUNITY TRIALS FOR THORACIC MALIGNANCIES 1. NSCLC Meta-Analysis Collaborative Group: Preoperative chemotherapy for non-small-cell lung cancer: A systematic review and metaanalysis of individual participant data. Lancet 383:1561-1571, 2014 2. Albain KS, Swann RS, Rusch VW, et al: Radiotherapy plus chemotherapy with or without surgical resection for stage III non-small-cell lung cancer: A phase III randomised controlled trial. Lancet 374:379-386, 2009 3. Betticher DC, Hsu Schmitz SF, Totsch M, et al: Mediastinal lymph node clearance after docetaxel-cisplatin neoadjuvant chemotherapy is prognostic of survival in patients with stage IIIA pN2 non-small-cell lung cancer: A multicenter phase II trial. J Clin Oncol 21: 1752-1759, 2003 4. Betticher DC, Hsu Schmitz SF, Totsch M, et al: Prognostic factors affecting long-term outcomes in patients with resected stage IIIA pN2 non-small-cell lung cancer: 5-Year follow-up of a phase II study. Br J Cancer 94: 1099-1106, 2006 5. Donington JS, Pass HI: Surgical approach to locally advanced non-small cell lung cancer. Cancer J 19:217-221, 2013 6. Lara-Guerra H, Waddell TK, Salvarrey MA, et al: Phase II study of preoperative gefitinib in clinical stage I non-small-cell lung cancer. J Clin Oncol 27:6229-6236, 2009 7. Rizvi NA, Rusch V, Pao W, et al: Molecular characteristics predict clinical outcomes: Prospective trial correlating response to the EGFR tyrosine kinase inhibitor gefitinib with the presence of sensitizing mutations in the tyrosine binding domain of the EGFR gene. Clin Cancer Res 17:3500-3506, 2011 8. Lynch TJ, Bell DW, Sordella R, et al: Activating mutations in the epidermal growth factor receptor underlying responsiveness of non–
330
9.
10.
11.
12.
13.
14.
15.
16.
small-cell lung cancer to gefitinib. N Engl J Med 350:2129-2139, 2004 Paez JG, Janne PA, Lee JC, et al: EGFR mutations in lung cancer: Correlation with clinical response to gefitinib therapy. Science 304:1497-1500, 2004 Altorki N, Lane ME, Bauer T, et al: Phase II proof-of-concept study of pazopanib monotherapy in treatment-naive patients with stage I/II resectable non-small-cell lung cancer. J Clin Oncol 28:3131-3137, 2010 Nikolinakos PG, Altorki N, Yankelevitz D, et al: Plasma cytokine and angiogenic factor profiling identifies markers associated with tumor shrinkage in early-stage non-small cell lung cancer patients treated with pazopanib. Cancer Res 70:2171-2179, 2010 Machtay M, Paulus R, Moughan J, et al: Defining local-regional control and its importance in locally advanced non-small cell lung carcinoma. J Thorac Oncol 7:716-722, 2012 Mauguen A, Pignon JP, Burdett S, et al: Surrogate endpoints for overall survival in chemotherapy and radiotherapy trials in operable and locally advanced lung cancer: A re-analysis of meta-analyses of individual patients' data. Lancet Oncol 14:619-626, 2013 Pazdur R: Endpoints for assessing drug activity in clinical trials. Oncologist 13:19-21, 2008 (suppl 2) Benveniste MF, Moran CA, Mawlawi O, et al: FDG PET-CT aids in the preoperative assessment of patients with newly diagnosed thymic epithelial malignancies. J Thorac Oncol 8: 502-510, 2013 Tsutani Y, Takuwa T, Miyata Y, et al: Prognostic significance of metabolic response by positron emission tomography after neoadjuvant chemotherapy for resectable malignant pleural mesothelioma. Ann Oncol 24:1005-1010, 2013
17. Lee HY, Lee KS, Park J, et al: Baseline SUVmax at PET-CT in stage IIIA non-small-cell lung cancer patients undergoing surgery after neoadjuvant therapy: Prognostic implication focused on histopathologic subtypes. Acad Radiol 19:440-445, 2012 18. Tsao AS, He D, Saigal B, et al: Inhibition of c-Src expression and activation in malignant pleural mesothelioma tissues leads to apoptosis, cell cycle arrest, and decreased migration and invasion. Mol Cancer Ther 6:1962-1972, 2007 19. William Jr. WN, Pataer A, Kalhor N, et al: Computed tomography RECIST assessment of histopathologic response and prediction of survival in patients with resectable non-smallcell lung cancer after neoadjuvant chemotherapy. J Thorac Oncol 8:222-228, 2013 20. Brahmer JR: Harnessing the immune system for the treatment of non-small-cell lung cancer. J Clin Oncol 31:1021-1028, 2013 21. Brahmer JR, Tykodi SS, Chow LQ, et al: Safety and activity of anti-PD-L1 antibody in patients with advanced cancer. N Engl J Med 366: 2455-2465, 2012 22. Brahmer J, Rizvi N, Lutzky J, et al: Clinical activity and biomarkers of MEDI4736, an antiPD-L1 antibody, in patients with NSCLC. J Clin Oncol ;32 [Abstract 8021] 23. Davies M: New modalities of cancer treatment for NSCLC: Focus on immunotherapy. Cancer Manage Res 6:63-75, 2014 24. Thomas A, Hassan R: Immunotherapies for non-small-cell lung cancer and mesothelioma. Lancet Oncol 13:e301-e310, 2012 25. Mansfield A, Peikert T, Roden A, et al: Programmed cell death 1 ligand 1 expression and association with survival in mesothelioma. J Thorac Oncol 9:S7-S52, 2014
Seminars in Thoracic and Cardiovascular Surgery Volume 26, Number 4