Virtual Library

Start Your Search

Kumar Prabhash

Moderator of

  • +

    OA11 - Decomplexifying Molecular Targets, Immunotherapy and Treatment Settings in the Real World (ID 137)

    • Event: WCLC 2019
    • Type: Oral Session
    • Track: Treatment in the Real World - Support, Survivorship, Systems Research
    • Presentations: 8
    • Now Available
    • +

      OA11.01 - Complex EGFR Mutations in Lung Adenocarcinoma (Now Available) (ID 2114)

      14:00 - 15:30  |  Presenting Author(s): Shang-Gin Wu  |  Author(s): Chong-Jen Yu, James Chih-Hsin Yang, Jin-Yuan Shih

      • Abstract
      • Presentation
      • Slides

      Background

      Epidermal growth factor receptor tyrosine kinase inhibitors (EGFR TKIs) provides a favorable treatment efficacy for EGFR-mutant lung cancer patients. Majority of EGFR mutations are a single mutation, including deletion in exon 19 (del-19) or L858R in exon 21. There is a subset of patients with complex EGFR mutations which contains two or more EGFR mutation types. It is unclear the treatment efficacy to different EGFR TKIs and survival prognosis for the complex EGFR-mutant patients due to small sample sizes of the prior studies. This study aimed to improve the understanding of the clinical characteristics and the prognosis of EGFR TKI treatment in lung adenocarcinoma patients with complex EGFR mutations.

      Method

      Between June 2005 to July 2018, patients harboring lung adenocarcinoma with complex EGFR mutations who were treated with EGFR TKIs were collected for EGFR mutation analysis by direct Sanger sequencing. Patients’ clinical characteristics, EGFR mutation status, treatment response, progression-free survival (PFS) and overall survival (OS) were analyzed. Patients harboring tumor with de novo T790M mutations were excluded for evaluation of EGFR TKI effectiveness.

      Result

      There were 175 patients (6.3%) with complex EGFR mutation from 2390 EGFR-mutant patients. Of the 175 complex EGFR-mutant patients, 122 patients who received EGFR TKIs were enrolled for evaluation of TKI effectiveness. Patients with the classical mutation pattern (del-19 or L858R) had higher treatment response rate (78.6% vs. 47.4%; p = 0.001) and PFS (8.6 months vs. 3.3 months; p = 0.006) than those without the classical mutations patterns (Fig-A). In multivariate analysis, female (p = 0.002), patients with disease relapse status, and the classical mutation patterns (p < 0.001) were associated with prolonged PFS. Compared with gefitinib and erlotinib, afatinib had a longer PFS, especially for patients without the classical mutation patterns. For OS, multivariate analysis revealed that female (p < 0.001), patients harbored classical mutation pattern (p = 0.001) (Fig-B), and patients with disease relapse status had longer OS. There were 51 patients who had re-biopsy tissue samples after acquired resistance to EGFR TKIs, 17 (33.3%) samples harbored T790M. In addition, small cell lung cancer transformation was detected in 3 (2%) patient’s re-biopsy tissue samples.

      layout 5.jpg

      Conclusion

      Female patients with complex EGFR-mutant lung adenocarcinoma and the classical mutation patterns have higher response rate, longer PFS and OS than those without the classical mutation patterns. Afatinib was active in lung adenocarcinoma harboring complex EGFR mutations, and may especially benefit patients without the classical mutation patterns due to longer PFS results.

      Only Members that have purchased this event or have registered via an access code will be able to view this content. To view this presentation, please login, select "Add to Cart" and proceed to checkout. If you would like to become a member of IASLC, please click here.

      Only Active Members that have purchased this event or have registered via an access code will be able to view this content. To view this presentation, please login or select "Add to Cart" and proceed to checkout.

    • +

      OA11.02 - Changes of Brain Structure in Advanced NSCLC Patients Receiving EGFR-TKIs: Dynamic Analysis Based on Series MRI Images (Now Available) (ID 314)

      14:00 - 15:30  |  Presenting Author(s): Chunli Luo  |  Author(s): Beisheng Yang, Xiaojuan Zhou, Lin Zhou, Ying Zhou, Jiang Zhu, Meijuan Huang, Feng Peng, Yongmei Liu, Yongsheng Wang, Zhiping Li, You Lu, Su Lui, Youling Gong

      • Abstract
      • Presentation
      • Slides

      Background

      EGFR-TKI was the standard care for metastatic NSCLC patients harboring positive EGFR mutation, which might inhibit EGF signaling pathway and consequently have effect on differentiation, maturation and rehabilitation of neural cells. For the first time, we evaluated the dynamic changes of white matter lesion (WML) and gray matter volume (GMV) among such patients based on series of MRI images.

      Method

      We retrospectively identified 778 patients with pathologically diagnosed advanced NSCLC receiving first-generation EGFR-TKIs in our hospital from 2010 to 2017, and 75 patients without brain metastasis and else comorbidity (hypertension, etc.) were analyzed. The modified Scheltens visual scale were performed to evaluate the changes of WML based on the series (baseline, 12 months' point and 24 months' point) of MRI images, and CBM (cluster-based morphometry) method based on SPM12 were adopted to identify GMV loss. The statistical methods were performed using SPSS software 22.0.

      Result

      During the 24-month EGFR-TKI treatment, the patient's WML visual scores showed a progressive worsen. Comparing to the baseline (6.683.636), the scores were significantly changed at the 12 months' point (8.650±3.857; Mean scores increasing 1.973, 95% CI 1.595-2.352, p<0.001) and changed more obviously at the 24 months' point (10.113.854; Mean scores increasing 3.427, 95% CI 2.979-3.874, p<0.001), respectively. Also, the significant GMV loss were found in subregions of the right occipital lobe (mean decrease 76.714, 95% CI 40.739-112.690), left occipital lobe (mean decrease 93.476, 95% CI 37.483-149.469) and left basal ganglia (mean decrease 37.571, 95% CI 21.576-53.567), respectively (all p<0.005, the cluster level FDR<0.05).

      Conclusion

      Dynamic structural analysis of series brain MRI images showed the significant worsen of the WML and GMV loss in patients with advanced NSCLC receiving EGFR-TKIs chronically. Perspective studies are warranted to verify its impact on the cognitive deficiency and hypomnesis among these patients in future.

      2019322fig1.jpg

      Only Members that have purchased this event or have registered via an access code will be able to view this content. To view this presentation, please login, select "Add to Cart" and proceed to checkout. If you would like to become a member of IASLC, please click here.

      Only Active Members that have purchased this event or have registered via an access code will be able to view this content. To view this presentation, please login or select "Add to Cart" and proceed to checkout.

    • +

      OA11.03 - Survival Disparities Between Academic and Community Centers in Advanced Lung Cancer in the US: Can We Bridge the Gap? (Now Available) (ID 1731)

      14:00 - 15:30  |  Presenting Author(s): Sendilnathan Ramalingam  |  Author(s): Michaela Dinan, Jeffrey Crawford

      • Abstract
      • Presentation
      • Slides

      Background

      Lung cancer causes the most cancer deaths in the US. Our prior study found widening 2-year survival (2YS) disparity between academic and community-based centers (ACs and CCs) prior to 2010, most apparent in adenocarcinoma, suggesting a treatment-related effect. We hypothesized this disparity continued to widen in more recent years.

      Method

      Retrospective study of outcomes through 2015 within the National Cancer Database. The primary outcome was 2YS. We used multivariable regression modeling, incorporating diagnosis year, facility type, age, gender, Charlson-Deyo score, and histology to compare AC/CC. A third facility type was identified, Integrated Network Cancer Programs (IC); we did a separate analysis incorporating AC/CC/IC. We formed six cohorts by 1)facility type and 2)time period (2004 through 2008, and 2011 through 2015). Hazard ratios were computed to compare survival between these six cohorts.

      Result

      98,069 patients were included. Treatment in ACs had superior 2YS compared to treatment in CCs, increasing from 16.1% versus 10.3% for those diagnosed in 2004(+5.8%), to 23.7% versus 16.2% for those diagnosed in 2013(+7.5%). Our multivariable model found growth in 2YS disparity of 0.34%-per-year (95% CI 0.18% to 0.50%, p<0.001). This was histology-related: Difference in adenocarcinoma 2YS rose from 7% in 2004 to 9% in 2013(p=0.0023), while squamous carcinoma 2YS difference was 2.7% in 2004 and 0.8% in 2013(p=0.6). 9047 patients were treated at ICs. In the 2004-2008 cohort treatment at ICs had similar outcomes to CCs, however by 2011-2015 ICs had superior histology-related survival, suggesting treatment-related improvements in ICs over CCs (Table 1).

      Conclusion

      Survival disparities in metastatic lung cancer between academic and community-based centers in the US continued to widen through 2015. Treatment at integrated centers, a group of facilities with at least one hospital that can include community and academic centers, may help to bridge the divide. Treatment related disparities in other health systems warrant further study globally.

      table 1.png

      Only Members that have purchased this event or have registered via an access code will be able to view this content. To view this presentation, please login, select "Add to Cart" and proceed to checkout. If you would like to become a member of IASLC, please click here.

      Only Active Members that have purchased this event or have registered via an access code will be able to view this content. To view this presentation, please login or select "Add to Cart" and proceed to checkout.

    • +

      OA11.04 - Discussant - OA11.01, OA11.02, OA11.03 (Now Available) (ID 3769)

      14:00 - 15:30  |  Presenting Author(s): Joel W Neal

      • Abstract
      • Presentation
      • Slides

      Abstract not provided

      Only Members that have purchased this event or have registered via an access code will be able to view this content. To view this presentation, please login, select "Add to Cart" and proceed to checkout. If you would like to become a member of IASLC, please click here.

      Only Active Members that have purchased this event or have registered via an access code will be able to view this content. To view this presentation, please login or select "Add to Cart" and proceed to checkout.

    • +

      OA11.05 - Optimizing Resources with Immunotherapy in Developing Countries: Experience in a Reference Center in Mexico (Now Available) (ID 2867)

      14:00 - 15:30  |  Presenting Author(s): Omar Eleazar Macedo-Perez  |  Author(s): Ivan Lyra Gonzalez, Leticia Bornstein, Katya Campos-Peralta, Jose Luis Aguilar-Ponce, Adan Gomez-Ponce, Angel Herrera-Gomez, Edgardo Jimenez-Fuentes

      • Abstract
      • Presentation
      • Slides

      Background

      Immunotherapy has proven clinical benefit in several tumors as a first line therapy or after standard treatment failure. Pivotal trials with immunotherapy were designed using a weight-based dose. However, recently as part of an effort to standardize the dose of the most common drugs (Nivolumab and Pembrolizumab) the Regulatory Agencies and Pharmaceutics have changed its prescription to fixed doses. In those pivotal studies, the median weight was around 70 to 80 kg. Our hypothesis estimates that a fixed dose increases the use of drug above effective thresholds increasing unnecessary expenses.

      Method

      We analyzed a cohort of patients treated with immunotherapy due to lung cancer (NSCLC) diagnosis either during first or second-line of treatment between 2016 and 2018 in the Thoracic Tumors Clinic at National Cancerology Institute in Mexico. We analyzed and estimated the median body in our population. Then, we compared treatment costs between weight-based doses (Nivolumab 3mg/Kg q.2 weeks and Pembrolizumab 2mg/Kg q.3 weeks) versus flat dose treatment (Nivolumab 240mg q.2 weeks and Pembrolizumab 200mg q.3 weeks).

      Result

      792 patients were included with and a median weight of 65 kg (SD ±13.11) was determined. Fixed dose of pembrolizumab administered during one year (17 applications) had an annual cost of $135,218 dlls. while the cost with weight-based dose was $87,913 dlls. with a net difference of $47,305 dlls. In the case of Nivolumab, an annual fixed-dose treatment (26 applications) has a cost of $114,816 dlls, while the weight-based dose has a cost of $93,392 dlls. with a net difference of $21,424 dlls (table1).

      PEMBROLIZUMAB (2 mg/Kg) Q3W vs 200mg Q3W

      Comparison

      Total dose per cycle

      Cost per cycle

      Excess cost per cycle

      Dose difference

      Annual cost treatment (17 cycles)

      Net difference annual cost per patient

      Weight based dose

      (Median 65Kg)

      130 mg

      $5171.4

      $2,782.6

      70mg

      $87,913

      $47,305

      Fixed-dose

      200mg

      $7,954

      NA

      NA

      $135,218

      NA

      NIVOLUMAB (3mg/Kg) Q2W vs 240mg Q2W

      Comparison

      Total dose per cycle

      Cost per cycle

      Excess cost per cycle

      Dose difference

      Annual cost treatment (26 cycles)

      Weight based dose

      (Median 65Kg)

      195mg

      $3,592

      $824

      45mg

      $93,392

      $21,424

      Fixed-dose

      240mg

      $4,416

      NA

      NA

      $114,816

      NA

      Conclusion

      General population in developing countries like Mexico are experiencing serious difficulties to get access to immunotherapy due to lack of coverage through Public Health Care System based in costs. According with our study, optimization of resources with weight-base dose could allow us to increase the rate of treated patients. Then, according with our analysis, in the case of Pembrolizumab we calculated coverage of 154 treatments instead of 100 using same budget and favoring the use of weight-based dose. While, in the case of Nivolumab we could increase the number of patients treated from 100 to 126 using weight-based dose. Therefore, our results support that therapies like immunotherapy should be calculated based on body weight as an attempt to increase access and avoid unnecessary expenses in Health Care Systems with limited resources.

      Only Members that have purchased this event or have registered via an access code will be able to view this content. To view this presentation, please login, select "Add to Cart" and proceed to checkout. If you would like to become a member of IASLC, please click here.

      Only Active Members that have purchased this event or have registered via an access code will be able to view this content. To view this presentation, please login or select "Add to Cart" and proceed to checkout.

    • +

      OA11.06 - Alternative Nivolumab (N) Duration and Scheduling in Advanced Non-Small Cell Lung Cancer (aNSCLC): Real-Life Data (Now Available) (ID 1921)

      14:00 - 15:30  |  Presenting Author(s): Elizabeth Dudnik  |  Author(s): Mor Tal Moskovitz, Abed Agbarya, Teodor Kuznetsov, Tzippy Shochat, Damien Urban, Mijana Wollner, Alona Zer, Ofer Rotem, Nir Peled, Jair Bar

      • Abstract
      • Presentation
      • Slides

      Background

      Little is known regarding the optimal scheduling and treatment (Tx) duration of N in aNSCLC. Stopping N after 1 year of Tx negatively affects outcomes.

      Method

      45 consecutive aNSCLC patients (pts) receiving N for ≥2 years (y) were identified in the electronic databases of 4 Israeli cancer centers. These were divided into Groups A (N continued for >2y at a dose 3mg/kg q2w/240mg q2w; n-21), B (N continued for >2y at a dose 3mg/kg q3w-q8w/480mg q4w; n-17), and C (N stopped at 2y for reason other than progressive disease or intolerable toxicity; n-7). PFS (RECIST 1.1) and safety since 2y after N initiation were assessed.

      Result

      Baseline, treatment characteristics and outcomes are presented in the Table and the Picture. Allocation to Group B and C was associated with HR for PFS-2.4 (95%CI, 0.3-18.8, p-0.4) and HR for PFS-3.3 (95%CI, 0.3-30.9, p-0.3), respectively. After 2y since N initiation, new N-related toxicity developed in 24%, 18%, and 28% of pts in Groups A, B, and C, respectively (p-NS).

      table wclc.jpgpicture wclc.jpg

      Conclusion

      A trend for worse outcomes was observed with alternative N scheduling/N quitting 2y after initiation. So far, continuing N at a standard dose until disease progression/ intolerable toxicity remains the standard treatment option.

      Only Members that have purchased this event or have registered via an access code will be able to view this content. To view this presentation, please login, select "Add to Cart" and proceed to checkout. If you would like to become a member of IASLC, please click here.

      Only Active Members that have purchased this event or have registered via an access code will be able to view this content. To view this presentation, please login or select "Add to Cart" and proceed to checkout.

    • +

      OA11.07 - Chemotherapy Plus EGFR-TKI as First-Line Treatment Provides Better Survival for EGFR Mutation NSCLC Patients: Update Data for NCT02148380 (Now Available) (ID 2207)

      14:00 - 15:30  |  Presenting Author(s): Wei Zhang  |  Author(s): Jianlin Xu, Yuqing Lou, Yanwei Zhang, Baohui Han

      • Abstract
      • Presentation
      • Slides

      Background

      Previously, we did a prospective study to compare pemetrexed plus carboplatin and gefitinib to either pemetrexed plus carboplatin or gefitinib alone as first-line therapy for lung adenocarcinoma patients harboring sensitive EGFR mutations (NCT02148380). The primary endpoint PFS was met at Oct 1, 2016. However, the OS of combinational group was not mature then [Han B, et al. Int J Cancer. 2017;141:1249-1256]. In the present study, we continued the OS follow-up until Sep 28 2018.

      Method

      The survival curves for OS were estimated with the Kaplan-Meier method and were compared between combination and gefitinib groups using the log-rank test. 2-years, 3-years survival rates were compared between combination and gefitinib groups using Pearson Chi-Square.

      Result

      Baseline characteristics of the intent-to-treat (ITT) population have been reported. At last day of follow-up (Sep 28 2018), 30 (75.0%) patients in the combinational group, 35 (85.4%) patients in the gefitinib group died. 2-year survival rates of combinational and gefitinib groups were 85.0% (34/40), 56.1% (23/41) (P=0.004), respectively. 3-year survival rates of combinational and gefitinib groups were 52.5% (21/40), 24.4% (10/41) (P=0.009), respectively. The median OS was 37.9 months (95%CI: 17.3-58.6) for the combinational group, which was substantially longer than the median OS for first-line gefitinib group (25.8 months [95%CI: 19.2-32.3]). The HR of combinational group versus gefitinib group was 0.56 (95%CI:0.34-0.91, P=0.02).

      19del: The median OS was 51.0 months (95%CI: 36.6-65.5) for the combinational group, which was substantially longer than the median OS for first-line gefitinib group (29.8 months [95%CI: 26.7-32.9]). The HR of combinational group versus gefitinib group was 0.61 (95%CI:0.30-1.25, P=0.18).

      21L858R: The median OS was 32.3 months (95%CI: 27.8-36.7) for the combinational group, which was substantially longer than the median OS for first-line gefitinib group (22.8 months [95%CI: 13.1-32.5]). The HR of combinational group versus gefitinib group was 0.50 (95%CI:0.25-1.00, P=0.05).

      Totally, 15 patients had baseline central nervous system (CNS) metastases. The median OS of patients who had baseline CNS metastases was 25.6 months (95%CI: 15.1-36.1); the median OS of patients who had no baseline CNS metastases was 31.7 months (95%CI: 28.2-35.2). The HR of CNS metastases group versus no CNS metastases group was 2.80 (95%CI:1.51-5.18, P=0.001). Among the combinational group, 20% (8/40) percent of patients had baseline CNS metastases. 17.1% (7/41) percent of patients in the gefitinib group had baseline CNS metastases.

      CNS: The median OS was 27.0 months, (95%CI: 21.8-32.3) for the combinational group, which was substantially longer than the median OS for first-line gefitinib group (15.5 months, 95%CI: 6.8-24.3). The HR of combinational group versus gefitinib group was 0.17 (95%CI:0.04-0.68, P=0.013).

      No CNS: The median OS was 47.4 months, 95%CI: 27.2-67.7 for the combinational group, which was substantially longer than the median OS for first-line gefitinib group (27.4 months, 95%CI: 23.0-33.7). The HR of combinational group versus gefitinib group was 0.57 (95%CI:0.32-0.99, P=0.044).

      Conclusion

      The current study on lung adenocarcinoma patients harboring sensitive EGFR mutations showed that the combined treatment with pemetrexed plus carboplatin with gefitinib provide better survival benefits than gefitinib alone.

      Only Members that have purchased this event or have registered via an access code will be able to view this content. To view this presentation, please login, select "Add to Cart" and proceed to checkout. If you would like to become a member of IASLC, please click here.

      Only Active Members that have purchased this event or have registered via an access code will be able to view this content. To view this presentation, please login or select "Add to Cart" and proceed to checkout.

    • +

      OA11.08 - Discussant - OA11.05, OA11.06, OA11.07 (Now Available) (ID 3770)

      14:00 - 15:30  |  Presenting Author(s): Ullas Batra

      • Abstract
      • Presentation
      • Slides

      Abstract not provided

      Only Members that have purchased this event or have registered via an access code will be able to view this content. To view this presentation, please login, select "Add to Cart" and proceed to checkout. If you would like to become a member of IASLC, please click here.

      Only Active Members that have purchased this event or have registered via an access code will be able to view this content. To view this presentation, please login or select "Add to Cart" and proceed to checkout.



Author of

  • +

    P1.01 - Advanced NSCLC (ID 158)

    • Event: WCLC 2019
    • Type: Poster Viewing in the Exhibit Hall
    • Track: Advanced NSCLC
    • Presentations: 1
    • Moderators:
    • Coordinates: 9/08/2019, 09:45 - 18:00, Exhibit Hall
    • +

      P1.01-88 - PS 2 Patients with Advanced EGFR Mutant NSCLC: Subset Analysis of a Phase III Randomized Trial Comparing Gefitinib to Gefitinib with Chemotherapy (ID 3022)

      09:45 - 18:00  |  Presenting Author(s): Kumar Prabhash

      • Abstract

      Background

      Background: Performance status is an important prognostic indicator in lung cancer. Most landmark trials excluded PS 2 patients; however, many patients we see in the clinic are PS 2, hence management of these patients is an extrapolation of the data from trials done in fitter patients.

      Method

      Methods: We had conducted a Phase III randomized trial in patients with advanced NSCLC harboring EGFR sensitizing mutation, ECOG PS 0 to 2 planned for first-line palliative therapy. Performance status was a stratification factor, thus there was an equal proportion of PS 2 patients in the two arms. Randomization was 1:1 to gefitinib 250 mg orally daily (gef) or pemetrexed 500 mg/m2 and carboplatin AUC 5 IV 3-weekly for 4 cycles, followed by maintenance pemetrexed with gefitinib from day 1 (gef+C). Primary end-point was progression-free survival (PFS); secondary end-points included overall survival (OS), response rate and toxicity. We present the subset analysis of the PS 2 patients in the trial.

      Result

      Results: Between 2016 and 2018, 75 PS 2 patients were randomly assigned to gef (n=39) and gef+C (n=36). The median age was 56 years, 63% were females, 23% had brain metastases and 50% had comorbidities. Median follow-up was 20 months (range, 7 to 28). Radiologic response rates were 83% and 67% in gef+C and gef arms respectively, P=0.117. Estimated median PFS was significantly longer with gef+C than gef [15 months, (95% CI, 10.6 to 19.4) versus 6 months (95% CI, 4.3 to 7.7); hazard ratio for disease progression or death, 0.57; 95% CI, 0.33 to 0.98; P=0.031] (Fig 2). Data for OS are immature. Clinically relevant > grade 3 toxicities occurred in 58% and 28% of patients in gef+C and gef arms respectively, P=0.008.

      Conclusion

      picture1.jpgConclusion: In EGFR mutant PS 2 patients, combining pemetrexed and carboplatin chemotherapy with gefitinib led to a significant PFS prolongation, with an increase in clinically relevant severe toxicities. PS 2 patients with driver mutations appear to benefit from treatment intensification.

  • +

    P2.01 - Advanced NSCLC (ID 159)

    • Event: WCLC 2019
    • Type: Poster Viewing in the Exhibit Hall
    • Track: Advanced NSCLC
    • Presentations: 1
    • Moderators:
    • Coordinates: 9/09/2019, 10:15 - 18:15, Exhibit Hall
    • +

      P2.01-102 - Outcome of Patients with EGFR Exon 19 Mutation in a Phase III Randomized Trial Comparing Gefitinib to Gefitinib with Chemotherapy (ID 3027)

      10:15 - 18:15  |  Presenting Author(s): Kumar Prabhash

      • Abstract

      Background

      In patients with EGFR sensitizing mutations, the subtype of the EGFR mutation impacts the clinical outcome and carries both prognostic and predictive value.

      Method

      We had conducted a Phase III randomized trial in patients with advanced NSCLC harboring EGFR sensitizing mutation, ECOG PS 0 to 2 planned for first-line palliative therapy. The type of sensitizing EGFR mutation (exon 19 versus others) was a stratification factor. Randomization was 1:1 to gefitinib 250 mg orally daily (gef) or pemetrexed 500 mg/m2 and carboplatin AUC 5 IV 3-weekly for 4 cycles, followed by maintenance pemetrexed with gefitinib from day 1 (gef+C). The primary end-point was progression-free survival (PFS); secondary end-points included overall survival (OS), response rate and toxicity. We present the subset analysis of the patients with exon 19 in-frame deletion.

      Result

      Between 2016 and 2018, 216 patients with EGFR exon 19 mutation were randomly assigned to gef (n=109) and gef+C (n=107). The median age was 54 years, 51% were males, 20% were PS 2 and 21% had brain metastases. Median follow-up was 17 months (range, 7 to 30). Radiologic response rates were 81% and 75% in gef+C and gef arms respectively, P=0.29. Estimated median PFS was significantly longer with gef+C than gef [17 months, (95% CI, 12.4 to 21.6) versus 8 months (95% CI, 6.8 to 9.3); hazard ratio for disease progression or death, 0.49; 95% CI, 0.35 to 0.69; P<0.001]. Data for OS are immature. Clinically relevant > grade 3 toxicities occurred in 55% and 23% of patients in gef+C and gef arms respectively, P<0.001.

      Conclusion

      In patients with exon 19 in-frame deletion, combining pemetrexed and carboplatin chemotherapy with gefitinib led to a significant PFS prolongation, with an increase in clinically relevant severe toxicities.