ASCO Issues Update on Frontline NSCLC Treatment in Wake of Immunotherapy Approvals
March 21, 2020 08:00pm
By Audrey Sternberg
Systemic anticancer therapy recommendations for patients with stage IV non–small cell lung cancer (NSCLC) depend on individual tumor histology, patient performance status, and driver oncogene biomarker status, which are most often anaplastic lymphoma kinase (ALK) translocation or epidermal growth factor receptor (EGFR) mutation.
1-4Approximately 85% of patients with stage IV NSCLC do not have acti- vating EGFR mutations or ALK translocations, requiring distinct first-line therapy recommendations. Although the discovery of driver mutations has facilitated the development of new targeted therapies leading to improved patient outcomes, patients with nonsquamous NSCLC who do not harbor these mutations do not benefit from these new therapies.
Bevacizumab (Avastin) is a vascular endothelial growth factor (VEGF) inhibitor and angiogenesis inhibitor that functions by stopping new blood vessel growth, which kills cancer cells. Specifically, bevacizumab is a humanized monoclonal antibody that targets and suppresses the binding of VEGF-A, which prevents pro-angiogenic activity.5
In October 2006, the FDA approved a labeling extension for bevacizumab administered in combination with carboplatin and paclitaxel for the initial systemic treatment of patients with unresectable, locally advanced, recurrent, or metastatic nonsquamous NSCLC.6This approval was based on findings of the phase III randomized, active-controlled, open-label ECOG E4599 trial of 878 patients with locally advanced, metastatic, or recurrent nonsquamous NSCLC comparing bevacizumab plus pemetrexed/carboplatin versus pemetrexed/carboplatin alone.7
In this study, patients receiving bevacizumab plus pemetrexed/carboplatin exhibited a significant improvement in overall survival (OS; 12.3 vs 10.3 months, respectively), progression-free survival (PFS; 6.2 vs 4.5 months), and response rates (35% vs 15%) compared with patients receiving pemetrexed/carboplatin alone. The addition of bevacizumab was also associated with increased toxic effects, particularly pulmonary hemorrhage and febrile neutropenia; these risks must be considered in the context of the survival benefit for patients with NSCLC.
National Comprehensive Cancer Network (NCCN) guidelines state that bevacizumab plus paclitaxel/carboplatin has an NCCN category 1 recommendation based on a high level of evidence, making it the first-line standard of care in advanced or metastatic nonsquamous NSCLC.1In general, guidelines recommend first-line therapy with a platinum-based doublet chemotherapy regimen using carboplatin or cisplatin combined with bevacizumab.1,2,8,9The results of a recent study of 1578 patients with nonsquamous stage IV NSCLC without known actionable mutations revealed that carboplatin plus pemetrexed was the most frequently used induction regimen, either with (16%) or without (26%) bevacizumab.8The next most frequently used regimen was carboplatin plus paclitaxel, with (11%) or without (14%) bevacizumab.
Results of the BEYOND trial, a placebo-controlled, phase III study in Chinese patients with advanced or recurrent nonsquamous NSCLC, also support the use of first-line carboplatin/ paclitaxel plus bevacizumab.10This trial enrolled 276 patients who were randomized to receive carboplatin/paclitaxel with or without bevacizumab to evaluate survival outcomes. Patients receiving carboplatin/paclitaxel plus bevacizumab exhibited significantly prolonged PFS and OS rates compared with patients receiving carboplatin/paclitaxel plus placebo. Objective response rates (ORRs) were also improved in patients receiving carboplatin/paclitaxel plus bevacizumab. Importantly, median PFS was significantly prolonged in patients without actionable driver mutations.
Results from a meta-analysis of phase II/III trials in which bevacizumab was added to platinum-based chemotherapy regimens further support these data, demonstrating significantly prolonged PFS and OS in patients receiving chemotherapy plus bevacizumab versus bevacizumab alone.11
In the international, multicenter, single-arm, phase II BELIEF trial, 109 patients with stage IIIB or stage IV EGFR-mutant (exon 19 deletion or L858R mutation) NSCLC, stratified by T790M expression, were treated with erlotinib (Tarceva) plus bevacizumab with a primary endpoint of PFS.12 Data showed that the overal median PFS was 13.2 months with a 12-month PFS rate of 55%. In the T790M-positive group, median PFS was 16.0 months, with a 12-month PFS rate of 68%. In the T790M-negative group, median PFS was 10.5 months, with a 12-month PFS rate of 48%.
NCCN guidelines also recommend that bevacizumab should be given until disease progression.1The results from AvaALL, an open-label, randomized phase III trial investigating standard-of-care therapy with or without continuous bevacizumab treatment beyond progression in 475 patients with NSCLC, were presented at the 2017 ASCO Annual Meeting.13This study reported 10% higher OS rates in patients receiving bevacizumab versus standard of care alone at 6, 12, and 18 months. Furthermore, PFS was longer in patients continuing bevacizumab after second disease progression and significantly longer after third disease progression compared with patients receiving standard of care alone. The ORR and disease control rates were also slightly improved in patients continuing bevacizumab. No additional safety signals were identified in this study.
It is important to caution against the use of any regimen with a high risk of thrombocytopenia and bleeding in combination with bevacizumab.1Bevacizumab should only be given as a single agent as maintenance if initially used with chemotherapy.
A Japanese study evaluating first-line bevacizumab in patients with NSCLC most frequently reported grade 3/4 neutropenia and hypertension.14Leukopenia, neutropenia, anemia, thrombocytopenia, and febrile neutropenia were severe grade 3/4 hematological toxicities that were reported. Severe grade 3/4 bevacizumab-related AEs included hypertension (30.4%), proteinuria (5.9%), thromboembolism (4.9%), and epistaxis (1%). These AE rates in the ECOG E4599 trial were reported at 7%, 3.1%, 0.2%, and 0.7%, respectively.7The greater rate of hypertension in the Japanese study was attributed to the higher incidence of comorbid hypertension at baseline.14The authors reported that most AEs were manageable, with no bevacizumab-related mortality.
A study investigating toxicities associated with second-line or later docetaxel plus bevacizumab therapy revealed that patients treated with the combination developed grade 3/4 neutropenia (100%) or febrile neutropenia (26.7%).15 .The phase IV SAiL and ARIES studies confirmed the safety profile of combined chemotherapy plus bevacizumab in daily practice.16,17Additionally, a meta-analysis revealed data supporting increased hypertension, neutropenia, febrile neutropenia, proteinuria, and hemorrhages with the use of added bevacizumab.11
Contraindications to bevacizumab include hypersensitivity to the drug and recent pulmonary hemorrhage/hemoptysis. Bevacizumab treatment should be discontinued if the patient develops a need for major surgery (28-day interval needed), a gastrointestinal perforation, uncontrollable hypertension, hypertensive crisis or hypertensive encephalopathy, posterior reversible encephalopathy syndrome, grade 4 proteinuria, grade 4 venous thromboembolic reactions, or grade 3/4 bleeding.
Pemetrexed/Cisplatin or Pemetrexed/Carboplatin
The standard of care for patients with advanced NSCLC who do not have EGFR mutations or ALK rearrangements is a platinum doublet (preferred to be pemetrexed-based) for 4 to 6 cycles. Patients with adenocarcinoma treated with first-line cisplatin/ pemetrexed demonstrated significantly improved OS compared with cisplatin/gemcitabine in the trial by Scagliotti et al (12.6 vs 10.9 months).18,19Recommended agents include platinum agents (carboplatin, cisplatin), taxanes (paclitaxel, docetaxel, albumin-bound paclitaxel), etoposide, vinorelbine, pemetrexed, and gemcitabine.1
Platinum combinations have led to a plateau in overall response rate (25%-35%), median survival (8-10 months), time to progression (4-6 months), 1-year survival rate (30%-40%), and 2-year survival rate (10%-15%) in fit patients.1Unfit patients of any age (ie, those with a performance status of 3 or 4) will not benefit from cytotoxic treatment if they do not have actionable driver mutations. Patients should be reassessed after 2 cycles followed by CT imaging with or without contrast every 2 to 4 cycles or when clinically indicated.
A recent phase II study assessed the safety and efficacy of pemetrexed/cisplatin induction chemotherapy followed by pemetrexed/cisplatin plus concurrent radiotherapy in 90 patients with locally advanced nonsquamous NSCLC.20At 1 year, PFS was 51.3%, with median PFS and OS of 10.6 and 26.2 months, respectively. Treatment-related enteritis led to 1 patient death. Four patients discontinued because of treatment-related AEs (TRAEs). Grade 3/4 AEs reported by more than 1 patient during induction chemotherapy included neutropenia and syncope. AEs reported during concurrent chemotherapy/radiotherapy included grade 3/4 esophagitis, neutropenia, and leukopenia.
A study evaluating the safety and efficacy of pemetrexed combined with cisplatin or carboplatin demonstrated a control rate of 66.7% and a median PFS of 5.5 months in patients with nonsquamous NSCLC.21Commonly reported AEs included anemia, leukopenia, and gastrointestinal response.
The POINTBREAK trial demonstrated that carboplatin/ pemetrexed/bevacizumab is an acceptable treatment option, reporting similar OS rates compared with a taxane-based regimen.22Furthermore, this study confirmed that pemetrexed-based regimens are less toxic than taxane-based regimens.
Considerations for Treatment With Chemotherapy
When selecting treatment regimens, it is important to consider the patient’s performance status, which is a measure of the impact of tumor symptoms and other pre-existing medical problems and comorbidities on a patient’s ability of self-care and daily function.23The most commonly used performance status scales are the Karnofsky scale (scale from 0 to 100) and the Eastern Cooperative Oncology Group Scale of Performance Status (ECOG PS).24,25
The ECOG PS is a 5-point scale (5 is the worst) that is based on symptom interference with daily activities and the proportion of waking hours spent in bed. Patients are classified as PS2 if they are ambulatory and capable of self-care, but restricted in physical activity and require bed rest for less than 50% of waking hours.
A meta-analysis demonstrated evidence for a survival benefit with cisplatin-based chemotherapy in patients with PS0 or PS1; no consistent data support the efficacy of platinum-based treatments for patients with PS2.26-28A European Experts Panel recommended the use of chemotherapy in patients with advanced NSCLC and PS2.23Based on available evidence, single-agent chemotherapy with vinorelbine, gemcitabine, or taxanes was preferred. Alternative options included carboplatin-based or low-dose cisplatin-based doublets.
Pembrolizumab (Keytruda), a programmed death receptor-1 (PD- 1) inhibitor, is an immune checkpoint inhibitor that improves antitumor immunity. Cohort G of KEYNOTE-021, a randomized, open-label, phase II study of 123 patients with nonsquamous NSCLC, evaluated the efficacy of adding pembrolizumab to platinum-doublet chemotherapy.29Of these patients, 55% receiving pembrolizumab plus chemotherapy achieved an objective response compared with 29% of patients receiving chemotherapy alone. TRAE rates were similar between groups. Patients receiving pembrolizumab plus chemotherapy most frequently reported grade 3 or worse TRAEs of anemia, decreased neutrophil count, acute kidney injury, decreased lymphocyte count, fatigue, neutropenia, sepsis, and thrombocytopenia. Pembrolizumab was discontinued in 10% of patients due to AEs. Three treatment-related deaths were reported overall, with 1 due to sepsis in the combination group.
These findings led to the accelerated FDA approval of pembrolizumab in combination with pemetrexed and carboplatin for the treatment of patients with previously untreated metastatic nonsquamous NSCLC in May 2017.30Importantly, positive programmed death-ligand receptor-1 (PD-L1) expression levels were not required for treatment; however, patients with 50% or higher PD-L1 expression who received chemotherapy with pembrolizumab had higher response rates compared with patients receiving chemotherapy alone (80% vs 35%).
Pembrolizumab can contribute to immune-mediated adverse reactions, including pneumonitis, hepatitis, colitis, endocrinopathies, and nephritis.31Due to the severity of these potential adverse reactions, when they arise pembrolizumab should be stopped or discontinued and corticosteroids should be administered when appropriate.
Pembrolizumab also has an NCCN category 1 recommendation as subsequent therapy for patients with metastatic non-squamous NSCLC and PD-L1 expression based on findings from KEYNOTE-010, KEYNOTE-001, and the FDA approval.32-34KEYNOTE-010 was a phase II/III trial evaluating pembrolizumab in patients with advanced nonsquamous NSCLC who had PD-L1 expression levels of at least 1%.32OS was significantly prolonged in patients receiving pembrolizumab compared with patients receiving docetaxel. Patients receiving pembrolizumab also reported fewer grade 3 to 5 TRAEs compared with docetaxel.
The phase I KEYNOTE-001 trial evaluated the safety and efficacy of pembrolizumab in patients with metastatic NSCLC.33Patients with 50% or greater PD-L1 expression levels had an increased response rate, PFS, and OS compared with the entire patient population. Serious grade 3 or higher toxicity was reported in less than 10% of patients.
Sequencing of Current Therapies
Bevacizumab is the frontline standard of care for nonsquamous NSCLC. It is the only anti-angiogenic agent approved for the first-line treatment of NSCLC in select patients. With the use of immune checkpoint inhibitors in the second-line setting, third-line options for patients with nonbiomarker, nonsquamous NSCLC include docetaxel, erlotinib (Tarceva), and pemetrexed.35-37It is important to note that pemetrexed can only be prescribed if it was not previously used either in the first-line setting or in maintenance therapy.
A numerically increased PFS was reported in the REVEL trial in patients receiving ramucirumab (Cyramza), an anti-angiogenesis agent, and docetaxel compared with patients receiving docetaxel alone.38Fourth-line treatment options include any agents not administered in previous lines.39Limiting the selection of these therapies, no clinical trials have investigated these agents in later lines of therapy. Clinical trials may be considered for patients with satisfactory performance status.
For patients receiving second-line and beyond (subsequent) therapy, NCCN guidelines recommend response assessments of known disease sites using CT with contrast every 6 to 12 weeks.1For patients without actionable driver mutations who have PS0 to PS2 who have disease progression during or following first-line therapy, recommendations for subsequent systemic therapy include nivolumab (Opdivo), pembrolizumab, atezolizumab (Tecentriq), docetaxel with or without ramucirumab, or gemcitabine, if not already given. Patients with nonsquamous NSCLC may be prescribed pemetrexed.
Recently, a therapy-sequencing disease model that approximates treatment outcomes was developed for patients with non-squamous NSCLC based on published trial data.40All patients were treatment-naïve and received first-line doublet chemotherapy or bevacizumab-based therapy. All bevacizumab-based first-line sequences achieved a total PFS of approximately 15 months. First-line sequences with pemetrexed/cisplatin achieved a total PFS between 12.6 and 12.8 months, with a slightly prolonged PFS when assuming for pemetrexed maintenance therapy. Overall, these findings support improved survival outcomes following a first-line bevacizumab-based combination, which is likely due to the possibility of an additional line of treatment.
Radiation, Surgery, and Other Localized Treatments
Patients with stage I or II disease generally have the best chance for cure through surgery.41Clinicians must determine whether patients can tolerate surgery, although some patients considered inoperable may be tolerant of sublobar resection and/or minimally invasive surgery.41-45There are 4 types of surgery commonly used to treat lung cancer: wedge resection, lobectomy, pneumonectomy, and sleeve resection.46A wedge resection removes a tumor and some surrounding normal tissue; if a larger amount of tissue is taken, this is referred to as a segmental resection. A lobectomy refers to the removal of an entire lobe of the lung. A pneumonectomy is a procedure wherein 1 lung is removed. In a sleeve resection, part of the bronchus is removed.
Surgical resection is often not performed in patients with severely compromised lung function because a lobectomy can further compromise their condition. A recent retrospective study enrolled 50 patients with a predicted postoperative carbon monoxide diffusing capacity of 40% or less.47These patients underwent a lobectomy and experienced no operative complications, a median hospital stay of 5 days, and a 5-year OS rate of 69%, suggesting that a lobectomy is feasible in individuals with poor lung function.
Radiation therapy (RT) may be used in all stages of NSCLC, including as definitive therapy for locally advanced NSCLC, generally combined with chemotherapy; definitive therapy for early-stage NSCLC in patients with contraindications for surgery; preoperative or postoperative therapy for selected patients treated with surgery; therapy for limited recurrences and metastases; and/or palliative therapy for patients with incurable NSCLC.1RT goals include maximization of tumor control and minimization of treatment toxicity. Radiation is given directly through an endoscope to tumors in the airways. External radiation therapy (ERT) is delivered via a machine outside the body. Internal RT is delivered through needles, seeds, wires, or catheters that are placed in or near the cancer.
RT is often administered in patients with inoperable disease due to comorbidities or poor performance status.48Performance status has been shown to determine outcomes in patients treated with conventional radiation.49Patients with PS0 or PS1 had significantly longer median OS compared with patients with PS2 or PS3 (23 vs 6 months) in a series of 62 patients with stage II NSCLC.50
Better outcomes have been reported with concurrent chemotherapy and radiation compared with sequential or single-modality therapy.51-53Nearly 25% of patients included in RTOG 9410 had a Karnofsky performance status of 70 to 80.53Patients with a poor performance status exhibited significantly worse OS compared with patients with a good performance status. In the CALGB 30106 study, which evaluated the role of the combination of chemoradiotherapy plus gefitinib (Iressa), an EGFR inhibitor, patients with a poor performance status had a significantly worse response rate than patients with a good performance status.54However, these patients with poor performance status had a better median PFS (13.4 vs 9.2 months) and OS (19.0 vs 13.0 months).
In the INT1039 study of patients with T1-3pN2M0 disease, the authors investigated the role of tri-modality therapy.55Of these patients, approximately 12% had a Karnofsky performance status of 70 to 80. This study did not support a role for performance status as an independent risk factor for survival. Based on these conflicting results, it is difficult to assess the benefit of standard therapy in patients with locally advanced disease and poor performance status. However, with concurrent chemotherapy and radiation, risks for toxicity, especially myelosuppression and esophagitis, can be higher. Patients with poor performance status are more likely to have chemotherapy-related complications secondary to myelosuppression, including neutropenia and anemia.56Therefore, sequential therapy use may be appropriate in this population, with the administration of systemic chemotherapy followed by RT.
Stereotactic body RT, a type of ERT, involves special equipment to ensure exact positioning of the patient for subsequent large doses of radiation aimed directly at the tumor, with less damage to nearby healthy tissue.46Stereotactic radiosurgery (SRS), another type of ERT, involves the use of a rigid head frame attached to the skull to immobilize the head during radiation delivery for treating lung cancer brain metastases.
Treatment for Central Nervous System (CNS) Metastases
In patients with newly diagnosed solitary brain metastasis, surgical resection and radiation are recommended to increase the lifespan in relatively young patients with good functional status.57For many years, whole-brain RT (WBRT) was the standard of care for patients with multiple NSCLC brain metastases.58Although WBRT can control brain disease in some patients, neurocognitive toxicity, including moderate to severe dementia occurring several months to years after treatment, has been observed in 2% to 5% of patients.59To avoid these toxicity issues, patients with a limited number of metastases and a better prognosis have been advised to undergo SRS alone, which has led to similar survival compared with initial treatment with WBRT and SRS.61-63Improved local and distant CNS control and decreased risk of death were reported in patients receiving adjuvant WBRT compared with observation only.64No survival benefits have been reported to date with adjuvant RT in the CNS.68In patients with no clinically actionable mutations and 1 to 3 CNS metastases without extracranial disease, chemotherapy and local CNS treatment are recommended. Patients with >3 CNS metastases, with or without extracranial disease and a plan to receive immunotherapy, should receive the immunotherapy treatment upfront and delay or omit WBRT.65
Current NCCN guideline recommendations for the treatment of limited brain metastases include SRS alone and surgical resection for select patients (those with symptomatic metastases or whose tumor tissue is needed for diagnosis) followed by SRS or WBRT.1Although patients with brain metastases were initially excluded from bevacizumab-based strategies due to the risk of bleeding into the brain, a retrospective review suggested that patients with CNS metastases have a similar risk of developing cerebral hemorrhage, independent of treatment with bevacizumab.66Recent data from phase II trials evaluating first-line carboplatin/paclitaxel with bevacizumab and second-line erlotinib/bevacizumab in patients with stage IV NSCLC and untreated brain metastases revealed a very low incidence (1/91) of nonfatal intracranial bleeding and encouraging PFS and OS outcomes and overall response rates.67
Based on ASCO guidelines, maintenance therapy is administered for patients with stable disease or those who respond to 4 cycles of chemotherapy.2Based on NCCN guidelines, patients who respond after 4 to 6 cycles may be administered maintenance therapy.1
Continuation maintenance employs the use of 1 or more agents used in rst-line induction therapy, whereas switch maintenance employs the use of a new agent. According to NCCN guidelines, bevacizumab and pemetrexed, either alone or in combination, or gemcitabine are recommended for continuation maintenance and pemetrexed or erlotinib is recommended for switch maintenance.1,2A recent study of patients with nonsquamous stage IV NSCLC without known actionable mutations reported that continuation of maintenance was only used in 26% of patients.8
The decision to administer maintenance therapy depends on several factors. Tumor histology is important in that pemetrexed is indicated only in patients with nonsquamous NSCLC.68Other factors include genomics, response to induction, patient health status (patients with PS2 or greater do not benefit), and patient choice.2The PARAMOUNT trial demonstrated that patients desire an OS benefit of at least 3 months or improved symptom control in order to balance the mild-to-moderate side effects.69
After completion of rst-line therapy, maintenance therapy is administered prior to disease progression. In the PARAMOUNT trial, patients with nonsquamous NSCLC who received peme- trexed maintenance after first-line chemotherapy exhibited significantly reduced disease progression compared with patients receiving placebo.69Furthermore, extended PFS and OS have been reported in patients receiving pemetrexed maintenance therapy.70Initially, the SATURN trial supported the administration of erlotinib for switch maintenance; however, the IUNO trial reported nonsigni cant results.71As a result, erlotinib is not an appropriate option for maintenance therapy in patients with non- biomarker, nonsquamous NSCLC.
AVAPERL, a randomized phase III trial, investigated the efficacy and safety of bevacizumab maintenance therapy with or without pemetrexed in 376 patients with nonsquamous NSCLC.72Patients received first-line cisplatin/pemetrexed plus bevacizumab every 3 weeks for 4 cycles. Patients achieving disease control (n = 253) were randomized to receive maintenance bevacizumab plus pemetrexed (n = 128) or bevacizumabalone (n = 125) until progression or consent withdrawal. At a median follow-up of 14.8 months, patients receiving bevacizumab plus pemetrexed exhibited significantly prolonged PFS compared with patients receiving bevacizumab alone. OS was numerically longer in patients receiving the combination compared with patients receiving bevacizumab alone.
A retrospective study had 160 Japanese patients with nonsquamous NSCLC receive first-line cisplatin/pemetrexed plus pembrolizumab followed by maintenance pembrolizumab to evaluate efficacy based on bevacizumab eligibility.73Patients eligible for bevacizumab exhibited significantly prolonged PFS and OS compared with patients ineligible for bevacizumab. These results indicate a favorable outcome with cisplatin/pemetrexed in a population similar to that of AVAPER, and further support use of the regimen from the PARAMOUNT trial for patients with advanced nonsquamous NSCLC, regardless of bevacizumab eligibility.
Although emerging evidence has supported the prolonged administration of pemetrexed in advanced NSCLC, most patients will experience disease progression during pemetrexed maintenance after a median PFS of 4 to 6 months.74In a report of 3 patients with advanced nonsquamous NSCLC, local control of thoracic oligoprogression was reported following stereotactic ablative radiotherapy during maintenance pemetrexed. Only mild radiation-related pneumonitis/fibrosis was reported.
The results of a meta-analysis of 87 records and 56 trials evaluating first-line therapy with maintenance in patients with advanced NSCLC demonstrated that for patients with no actionable driver mutations, first-line chemotherapy plus bevacizumab with bevacizumab maintenance was the only regimen that resulted in PFS and OS benefits that were clinically meaningful.75Specifically regarding nonsquamous NSCLC, a clinically meaningful benefit was observed with standard chemotherapy without maintenance. Meaningful PFS benefits were reported with first-line chemotherapy plus bevacizumab and chemotherapy plus bevacizumab plus dulanermin.
Patients with nonsquamous NSCLC and no actionable driver mutations have specific treatment recommendations. Positive results from recent studies have led to improved treatment options for this patient population. Regarding the recent approval of pembrolizumab, Martin Reck, MD, of Lung Clinic in Grosshansdorf, Germany, stated at the 2016 ESMO Annual Congress, “This data will completely change the management of patients with advanced NSCLC. This is primarily an opportunity for patients without oncogenic alterations.”76
Additionally, Johan Vansteenkists, MD, of the University Hospital KU Leuven, Belgium, commented, “A study is needed to confirm these findings in patients with high PD-L1 expression. Additional research should be done to find out whether patients with lower levels of PD-L1 expression also benefit more from pembrolizumab than chemotherapy.”