Outcomes of Rituximab-abbs versus Rituximab in Patients with Diffuse Large B-Cell Lymphoma in a Noninferiority Study.

BACKGROUND: Rituximab with cyclophosphamide, doxorubicin, vincristine, and prednisone (R-CHOP) is one of the first line treatments for diffuse large B-cell lymphoma (DLBCL). Rituximab comprises most of the treatment cost for this regimen; therefore, biosimilars, such as rituximab-abbs are crucial to provide affordable care. Although rituximab-abbs was studied primarily in follicular lymphoma, the Food and Drug Administration (FDA) approved this drug for all indications of the reference product on the basis of extrapolation. Effectiveness and safety data surrounding the use of rituximab-abbs in DLBCL is lacking. OBJECTIVE: To evaluate the effectiveness and safety of rituximab-abbs and reference product rituximab as R-CHOP treatment for patients with DLBCL. PATIENTS AND METHODS: This noninferiority (NI) study compared the 2-year overall survival (OS), overall response rate (ORR), and incidence of adverse events (AEs) between rituximab-abbs and its reference product (RP) in R-CHOP among adult patients with newly diagnosed DLBCL. The study inclusion period was from 1 January 2019 to 31 December 2020. Analyses were performed on the basis of a noninferiority lower limit of 10% for OS and ORR, and an upper limit of 10% for serious AEs. RESULTS: There were 240 patients who received RP rituximab, while 295 patients received rituximab-abbs. The cohort had a mean age of 63.7±12.2 years and 43% were female. The 2-year OS was 81.0% and 79.6% (NI p < 0.01) while the ORR was 80.0% and 69.6% (NI p < 0.01), among the rituximab-abbs and rituximab groups, respectively. The incidence of infusion reaction AEs (NI p < 0.01) and noninfusion reaction AEs (NI p < 0.01) also met noninferiority. CONCLUSIONS: We demonstrated that rituximab-abbs was noninferior to rituximab in both effectiveness and safety among patients receiving R-CHOP for DLBCL in this study. Long-term follow-up would be needed to confirm these results.

Vaccination of Adults With Cancer: ASCO Guideline.

PURPOSE: To guide the vaccination of adults with solid tumors or hematologic malignancies. METHODS: A systematic literature review identified systematic reviews, randomized controlled trials (RCTs), and nonrandomized studies on the efficacy and safety of vaccines used by adults with cancer or their household contacts. This review builds on a 2013 guideline by the Infectious Disease Society of America. PubMed and the Cochrane Library were searched from January 1, 2013, to February 16, 2023. ASCO convened an Expert Panel to review the evidence and formulate recommendations. RESULTS: A total of 102 publications were included in the systematic review: 24 systematic reviews, 14 RCTs, and 64 nonrandomized studies. The largest body of evidence addressed COVID-19 vaccines. RECOMMENDATIONS: The goal of vaccination is to limit the severity of infection and prevent infection where feasible. Optimizing vaccination status should be considered a key element in the care of patients with cancer. This approach includes the documentation of vaccination status at the time of the first patient visit; timely provision of recommended vaccines; and appropriate revaccination after hematopoietic stem-cell transplantation, chimeric antigen receptor T-cell therapy, or B-cell-depleting therapy. Active interaction and coordination among healthcare providers, including primary care practitioners, pharmacists, and nursing team members, are needed. Vaccination of household contacts will enhance protection for patients with cancer. Some vaccination and revaccination plans for patients with cancer may be affected by the underlying immune status and the anticancer therapy received. As a result, vaccine strategies may differ from the vaccine recommendations for the general healthy adult population vaccine.Additional information is available at www.asco.org/supportive-care-guidelines.

International Delphi Consensus on the Management of AQP4-IgG+ NMOSD: Recommendations for Eculizumab, Inebilizumab, and Satralizumab.

BACKGROUND AND OBJECTIVES: Neuromyelitis optica spectrum disorder (NMOSD) is a rare debilitating autoimmune disease of the CNS. Three monoclonal antibodies were recently approved as maintenance therapies for aquaporin-4 immunoglobulin G (AQP4-IgG)-seropositive NMOSD (eculizumab, inebilizumab, and satralizumab), prompting the need to consider best practice therapeutic decision-making for this indication. Our objective was to develop validated statements for the management of AQP4-IgG-seropositive NMOSD, through an evidence-based Delphi consensus process, with a focus on recommendations for eculizumab, inebilizumab, and satralizumab. METHODS: We recruited an international panel of clinical experts in NMOSD and asked them to complete a questionnaire on NMOSD management. Panel members received a summary of evidence identified through a targeted literature review and provided free-text responses to the questionnaire based on both the data provided and their clinical experience. Responses were used to generate draft statements on NMOSD-related themes. Statements were voted on over a maximum of 3 rounds; participation in at least 1 of the first 2 rounds was mandatory. Panel members anonymously provided their level of agreement (6-point Likert scale) on each statement. Statements that failed to reach a predefined consensus threshold (≥67%) were revised based on feedback and then voted on in the next round. Final statements were those that met the consensus threshold (≥67%). RESULTS: The Delphi panel comprised 24 experts, who completed the Delphi process in November 2021 after 2 voting rounds. In round 1, 23/25 statements reached consensus and were accepted as final. The 2 statements that failed to reach consensus were revised. In round 2, both revised statements reached consensus. Twenty-five statements were agreed in total: 11 on initiation of or switching between eculizumab, inebilizumab, and satralizumab; 3 on monotherapy/combination therapy; 7 on safety and patient population considerations; 3 on biomarkers/patient-reported outcomes; and 1 on research gaps. DISCUSSION: An established consensus method was used to develop statements relevant to the management of AQP4-IgG-seropositive NMOSD. These international statements will be valuable for informing individualized therapeutic decision-making and could form the basis for standardized practice guidelines.

Twelve tips for optimising learning for postgraduate doctors in the operating theatre.

Learning in the operating theatre forms a critical part of postgraduate medical education. Postgraduate doctors present a diverse cohort of learners with a wide range of learning needs that will vary by their level of experience and curriculum requirements. With evidence of both trainee dissatisfaction with the theatre learning experience and reduced time spent in the operating theatre, which has been exacerbated by the effects of the Covid-19 pandemic, it is vital that every visit to the operating theatre is used as a learning opportunity. We have devised 12 tips aimed at both learners and surgeons to optimise learning in the operating theatre, set out into four domains: educational context, preparation, learning in theatre, feedback and reflection. These tips have been created by a process of literature review and acknowledgment of established learning theory, with further discussion amongst surgical trainees, senior surgical faculty, surgical educators and medical education faculty.

Normal B-cell ranges in infants: A systematic review and meta-analysis.

BACKGROUND: During the first year of life, B-cell level is a valuable indicator of whether external factors, such as exposure to B-cell-depleting therapies, have an adverse impact on immune system development. However, there are no standard reference ranges of B-cell levels in healthy infants by age. OBJECTIVE: Our aim was to estimate the normal range of B-cell levels in infants, by age, during the first year of life by pooling data from published studies. METHODS: Studies reporting B-cell levels measured by using flow cytometry and CD19 markers in healthy infants were identified via a systematic literature review. Quality and feasibility assessments determined suitability for inclusion in meta-analyses by age group and/or continuous age. Means and normal ranges (2.5th-97.5th percentile) were estimated for absolute and percentage B-cell levels. Sensitivity analyses assessed the impact of various assumptions. RESULTS: Of the 37 relevant studies identified, 28 were included in at least 1 meta-analysis. The means and normal ranges of B-cell levels were found to be 707 cells/µL in cord blood (range 123-2324 cells/µL), 508 cells/µL in infants aged 0 to 1 month (range 132-1369 cells/µL), 1493 cells/µL in infants aged 1 to 6 months (range 416-3877 cells/µL), and 1474 cells/µL in infants older than 6 months (range 416-3805 cells/µL). The continuous age model showed that B-cell levels peaked at week 26. Trends were similar for the percentage B-cell estimates and in sensitivity analyses. CONCLUSION: These meta-analyses provide the first normal reference ranges for B-cell levels in infants, by week of age, during the first year of life.

Impact of short stature on quality of life: A systematic literature review.

OBJECTIVE: We sought to obtain a better understanding of the burden of short stature using a systematic literature review. METHODS: Studies of the burden of short stature, of any cause in adults and children, were searched using Embase, MEDLINE and Cochrane databases in April 2020, capturing publications from 2008 onwards. Case series and populations with adult-onset growth hormone deficiency (GHD) were excluded. RESULTS: Of 1684 publications identified, 41 studies (33 in children, 8 in adults) were included. All studies assessed human burden. Most study populations in children included short stature due to GHD, idiopathic short stature (ISS) and short stature after being born small for gestational age (SGA). In these populations, four studies showed that quality of life (QoL) in children with short stature was significantly worse than in children with normal stature. A significant association between QoL and short stature was observed in children with chronic kidney disease (CKD) (3 studies), achondroplasia (1 study) and transfusion-dependent ß-thalassaemia (1 study), and in samples with mixed causes of short stature (3 studies). Three studies (one in GHD/ISS/SGA and two in CKD) found no significant association between short stature and QoL, and several studies did not report statistical significance. Approximately half of adult studies showed that QoL was reduced with short stature, and the other half showed no association. Two studies, one in adults with Prader-Willi syndrome and one in children with GHD, suggested a potential association between short stature and poorer cognitive outcomes. Three studies demonstrated an increased caregiver burden in parents of children with short stature. CONCLUSIONS: Evidence suggests that, compared with those with normal stature, children and adults with short stature of any cause may experience poorer QoL. Further research could extend our understanding of the human burden in this field.

Regionalization of Acute Myeloid Leukemia Treatment in a Community-Based Population: Implementation and Early Results.

INTRODUCTION: Regionalization of care for acute myeloid leukemia (AML) has not been described for community-based settings. In 2015, we shifted AML induction from 21 local centers to 3 regional centers. METHODS: Using time-specific inception cohorts, we assessed whether regionalization was associated with the frequency of use of induction therapy, receipt of bone marrow transplantation, 60-day mortality (treatment toxicity), and 180-day mortality (treatment effectiveness). Information for all adult patients diagnosed with AML from 2013 to 2017 was obtained from the electronic health record. Multivariable methods were used to estimate the adjusted associations of induction, bone marrow transplantation, and death in relation to year of diagnosis before and after regionalization. RESULTS: Of 661 patients diagnosed during 2013 to 2017, 53% were ≥ 70 years, 22% were ≥ 80 years, and 10% died within the week following diagnosis. Comparing 2017 with 2013, the proportion of patients who received induction therapy increased 2.88 times (95% confidence interval [CI] = 1.55-5.35), and the proportion of non-acute promyelocytic leukemia patients receiving bone marrow transplantation increased 2.00 times (95% CI = 0.89-4.50). Regionalization was associated with lower 180-day mortality (hazard ratio [HR] = 0.64; 95% CI = 0.44-0.92), whereas change in 60-day mortality was not statistically significant (HR = 0.67; 95%CI = 0.43-1.04). CONCLUSION: In this community-based population, many patients were of advanced age yet benefitted from AML induction therapy delivered at a regionally specialized center. These early results suggest the benefit of regionalizing subspecialty leukemia care.

Atypical manifestations of coronavirus disease 2019 (COVID-19)-associated autoimmune thrombotic thrombocytopenic purpura.

Patients with coronavirus disease 2019 (COVID-19) infection can have various abnormal hematologic parameters. This report illustrates a case with unusual presentation of COVID-19-associated thrombotic thrombocytopenic purpura, in which the patient did not develop any typical respiratory signs or symptoms.

Development and Evaluation of a New Formative Assessment of Surgical Performance.

BACKGROUND: Formative assessment of operative performance is a mandatory part of surgical training. Engagement with assessment is limited in part by the time-consuming nature and the high perceived stakes of current assessment tools. OBJECTIVES: Our aims were to develop and collect validity evidence for a new operative assessment tool that addresses barriers to assessment that current trainers and trainees experience. METHODS: We developed the Generic Operative Supervised Learning Event (GOSLE). Orthopedic trainees were invited to complete GOSLEs with their trainers after surgical procedures. Experienced consultants assessed videotaped operations performed by trainees using the GOSLE. Validity evidence for content, relationships to other scores, internal structure, response process, and consequences of testing were evaluated. RESULTS: A total of 250 GOSLEs were completed. A strong correlation was found between the GOSLE scores and the Procedure-Based Assessment ratings (r = 0.87, p < 0.001). Rasch analysis confirmed satisfactory internal structure of the rating scale, with sequential increases in rating as performance improved. The reproducibility coefficient was 0.88, with 10 assessments of the same trainee who has to achieve a reliability coefficient of 0.8. Over 90% of users found the GOSLE easy to use, with most preferring it to other assessment methods. Feedback quality was higher using the GOSLE than with current assessments. CONCLUSION: We have collected validity evidence across multiple domains in support of the GOSLE. Its psychometric performance is comparable to that of current assessments. It is preferred by trainers and trainees over existing assessments. It stimulates high-quality, actionable feedback which better supports formative assessment. By addressing issues experienced with existing assessments, we expect engagement among users to be high.

Does co-payment for inhaler devices affect therapy adherence and disease outcomes? A historical, matched cohort study.

BACKGROUND: Adherence to asthma and chronic obstructive pulmonary disease (COPD) treatment has been shown to depend on patient-level factors, such as disease severity, and medication-level factors, such as complexity. However, little is known about the impact of prescription charges - a factor at the health care system level. This study used real-life data to investigate whether co-payment affects adherence (implementation and persistence) and disease outcomes in patients with asthma or COPD. METHODS: A matched, historical cohort study was carried out using two UK primary care databases. The exposure was co-payment for prescriptions, which is required for most patients in England but not in Scotland. Two comparison cohorts were formed: one comprising patients registered at general practices in England and the other comprising patients registered in Scotland. Patients aged 20-59 years with asthma, or 40-59 years with COPD, who were initiated on fluticasone propionate/salmeterol xinafoate, were included, matched to patients in the opposite cohort, and followed up for 1 year following fluticasone propionate/salmeterol xinafoate initiation. The primary outcome was good adherence, defined as medication possession ratio ≥80%, and was analyzed using conditional logistic regression. Secondary outcomes included exacerbation rate. RESULTS: There were 1,640 patients in the payment cohort, ie, England (1,378 patients with asthma and 262 patients with COPD) and 619 patients in the no-payment cohort, ie, Scotland (512 patients with asthma and 107 patients with COPD). The proportion of patients with good adherence was 34.3% and 34.9% in the payment and no-payment cohorts, respectively, across both disease groups. In a multivariable model, no difference in odds of good adherence was found between the cohorts (odds ratio, 1.04; 95% confidence interval, 0.85-1.27). There was also no difference in exacerbation rate. CONCLUSION: There was no difference in adherence between matched patients registered in England and Scotland, suggesting that prescription charges do not have an impact on adherence to treatment.

Inhaler Errors in the CRITIKAL Study: Type, Frequency, and Association with Asthma Outcomes.

BACKGROUND: Poor inhaler technique has been linked to poor asthma outcomes. Training can reduce the number of inhaler errors, but it is unknown which errors have the greatest impact on asthma outcomes. OBJECTIVE: The CRITical Inhaler mistaKes and Asthma controL study investigated the association between specific inhaler errors and asthma outcomes. METHODS: This analysis used data from the iHARP asthma review service-a multicenter cross-sectional study of adults with asthma. The review took place between 2011 and 2014 and captured data from more than 5000 patients on demographic characteristics, asthma symptoms, and inhaler errors observed by purposefully trained health care professionals. People with asthma receiving a fixed-dose combination treatment with inhaled corticosteroids and long-acting beta agonist were categorized by the controller inhaler device they used-dry-powder inhalers or metered-dose inhalers: inhaler errors were analyzed within device cohorts. Error frequency, asthma symptom control, and exacerbation rate were analyzed to identify critical errors. RESULTS: This report contains data from 3660 patients. Insufficient inspiratory effort was common (made by 32%-38% of dry-powder inhaler users) and was associated with uncontrolled asthma (adjusted odds ratios [95% CI], 1.30 [1.08-1.57] and 1.56 [1.17-2.07] in those using Turbohaler and Diskus devices, respectively) and increased exacerbation rate. In metered-dose inhaler users, actuation before inhalation (24.9% of patients) was associated with uncontrolled asthma (1.55 [1.11-2.16]). Several more generic and device-specific errors were also identified as critical. CONCLUSIONS: Specific inhaler errors have been identified as critical errors, evidenced by frequency and association with asthma outcomes. Asthma management should target inhaler training to reduce key critical errors.

Cost comparison of asthma treatments in 12-week study: caution about matching and short observational follow-up.

In the absence of randomisation, observational studies must take extra care to create treatment groups that are comparable in terms of key characteristics. Various matching methods exist which can create sound comparisons, minimising confounding where possible. A recent observational study by Dal Negro et al. carried out a cost analysis comparing two asthma medications. They report strong conclusions which favour one treatment over the other, however they include little discussion on the limitations of their study. The purpose of this letter is to comment on the weaknesses of the study design, including the level of matching used, and to urge readers to consider these issues alongside the interpretation of results.

Use of an embedded, micro-randomised trial to investigate non-compliance in telehealth interventions.

BACKGROUND/AIMS: Many types of telehealth interventions rely on activity from the patient in order to have a beneficial effect on their outcome. Remote monitoring systems require the patient to record regular measurements at home, for example, blood pressure, so clinicians can see whether the patient's health changes over time and intervene if necessary. A big problem in this type of intervention is non-compliance. Most telehealth trials report compliance rates, but they rarely compare compliance among various options of telehealth delivery, of which there may be many. Optimising telehealth delivery is vital for improving compliance and, therefore, clinical outcomes. We propose a trial design which investigates ways of improving compliance. For efficiency, this trial is embedded in a larger trial for evaluating clinical effectiveness. It employs a technique called micro-randomisation, where individual patients are randomised multiple times throughout the study. The aims of this article are (1) to verify whether the presence of an embedded secondary trial still allows valid analysis of the primary research and (2) to demonstrate the usefulness of the micro-randomisation technique for comparing compliance interventions. METHODS: Simulation studies were used to simulate a large number of clinical trials, in which no embedded trial was used, a micro-randomised embedded trial was used, and a factorial embedded trial was used. Each simulation recorded the operating characteristics of the primary and secondary trials. RESULTS: We show that the type I error rate of the primary analysis was not affected by the presence of an embedded secondary trial. Furthermore, we show that micro-randomisation is superior to a factorial design as it reduces the variation caused by within-patient correlation. It therefore requires smaller sample sizes - our simulations showed a requirement of 128 patients for a micro-randomised trial versus 760 patients for a factorial design, in the presence of within-patient correlation. CONCLUSION: We believe that an embedded, micro-randomised trial is a feasible technique that can potentially be highly useful in telehealth trials.

Design of telehealth trials--introducing adaptive approaches.

BACKGROUND: The field of telehealth and telemedicine is expanding as the need to improve efficiency of health care becomes more pressing. The decision to implement a telehealth system is generally an expensive undertaking that impacts a large number of patients and other stakeholders. It is therefore extremely important that the decision is fully supported by accurate evaluation of telehealth interventions. OBJECTIVE: Numerous reviews of telehealth have described the evidence base as inconsistent. In response they call for larger, more rigorously controlled trials, and trials which go beyond evaluation of clinical effectiveness alone. The aim of this paper is to discuss various ways in which evaluation of telehealth could be improved by the use of adaptive trial designs. RESULTS: We discuss various adaptive design options, such as sample size reviews and changing the study hypothesis to address uncertain parameters, group sequential trials and multi-arm multi-stage trials to improve efficiency, and enrichment designs to maximise the chances of obtaining clear evidence about the telehealth intervention. CONCLUSION: There is potential to address the flaws discussed in the telehealth literature through the adoption of adaptive approaches to trial design. Such designs could lead to improvements in efficiency, allow the evaluation of multiple telehealth interventions in a cost-effective way, or accurately assess a range of endpoints that are important in the overall success of a telehealth programme.

High-dose carmustine, etoposide, and cyclophosphamide followed by allogeneic hematopoietic cell transplantation for non-Hodgkin lymphoma.

Allogeneic hematopoietic cell transplantation (HCT) has been shown to be curative in a group of patients with aggressive non-Hodgkin lymphoma (NHL). A previous study has demonstrated equivalent outcomes with a conditioning regimen based on total body irradiation and another not based on total body irradiation with preparative therapy using cyclophosphamide, carmustine, and etoposide (CBV) in autologous HCT. We investigated the safety and efficacy of using CBV in an allogeneic setting. Patients were required to have relapsed or be at high risk for subsequent relapse of NHL. All patients had a fully HLA-matched sibling donor. Patients received carmustine (15 mg/kg), etoposide (60 mg/kg), and cyclophosphamide (100 mg/kg) on days -6, -4, and -2, respectively, followed by allogeneic HCT. All patients were treated with cyclosporine and methylprednisolone as prophylaxis for graft-versus-host disease (GVHD). Thirty-one patients (median age, 46 years) who were felt to be inappropriate candidates for autologous transplantation were enrolled. Each subject had a median of 3 previous chemotherapy regimens. All patients engrafted. Fifteen of 31 patients are alive. Median follow-up time was 11.5 months (range, .4-126). There were 8 deaths due to relapse. Nonrelapse mortality (n = 8) included infection (n = 3), GVHD (n = 2), diffuse alveolar hemorrhage (n = 1), veno-occlusive disease in the setting of concurrent acute GVHD of the liver (n = 1), and leukoencephalopathy (n = 1). Probabilities of event-free survival and overall survival were, respectively, 44% (95% confidence interval, 26%-62%) and 51% (33%-69%) at 1 year and 44% (26%-62%) and 47% (29%-65%) at 5 years. Probability of relapse was 33% (15%-51%) at 1 year and 5 years. Probability of nonrelapse mortality was 31% (13%-49%) at 1 year and 5 years. Incidences were 29% for acute GVHD and 39% for chronic GVHD. None of the 12 patients who developed chronic GVHD has disease recurrence. Patients who had required >3 previous chemotherapy regimens before HCT had an increased probability of relapse. CBV is an effective preparative regimen for patients with aggressive NHL who undergo allogeneic HCT.

Intermittent dosing of the farnesyl transferase inhibitor tipifarnib (R115777) in advanced malignant solid tumors: a phase I California Cancer Consortium Trial.

Tipifarnib (R115777) inhibits farnesylation of key proteins that modulate signaling pathways implicated in cell growth and proliferation, including members of the Ras and Rho families. It has broad-spectrum antiproliferative activity in vitro and in vivo. Clinical trials employing a continuous administration schedule have demonstrated dose-limiting neurotoxicity and myelosuppression. Preclinical studies have shown that intermittent oral administration can suppress tumor growth comparable to continuous administration. We conducted a National Cancer Institute-sponsored phase I trial to determine the feasibility of an intermittent dosing schedule of R115777 given orally twice daily on weeks 1 and 3 of a 28-day cycle in patients with malignant solid tumors. Starting dose was 300 mg twice daily (b.i.d.) with escalation by 300 mg b.i.d. increments over six dose levels to a maximum of 1800 mg b.i.d. Dose-limiting toxicity (DLT) was defined as any grade 3 or 4 non-hematologic toxicity, grade 4 thrombocytopenia, grade 4 neutropenia (ANC) with fever (38.3 degrees C or above) or a documented infection. Twenty-one patients with advanced solid tumors, all of whom had prior systemic therapy, were accrued. Grade 3 fatigue was dose limiting for two of three patients at the 900 mg b.i.d. dose level. Although no responses were seen, four of six patients with stable disease remained on study for at least a year (16, 17, 13 and 12 months) before developing progressive disease. Three of these prolonged stable disease patients had non-small cell lung cancer. We conclude that intermittent dosing of R115777 is feasible and tolerable. The recommended phase II dose is 600 mg orally b.i.d. on alternate weeks.

Gemcitabine in patients with non-small-cell lung cancer previously treated with platinum-based chemotherapy: a phase II California cancer consortium trial.

A phase II trial was designed to evaluate the efficacy and toxicity of gemcitabine in patients with non-small-cell lung cancer (NSCLC) previously treated with platinum-containing regimens and prospectively categorized for platinum response status. Treatment consisted of gemcitabine 1000 mg/m2 given intravenously on days 1 and 8 of a 21-day cycle. The status of p53 in pretreatment tumor tissue was assessed by immunohistochemistry (IHC). Sixty-one patients who progressed or recurred following platinum-based therapy were enrolled, 26 platinum-sensitive and 35 platinum-refractory. A median of 4 treatment courses (range, 2-7 courses) was delivered. Of the 55 patients assessable for response, there was 1 confirmed complete response and 3 with a confirmed partial response for an overall response proportion of 7%. Twenty-one patients had stable disease while 28 progressed and 2 patients had an unconfirmed partial response. Three of the responders (2 confirmed, 1 unconfirmed) were platinum-refractory. Median progression-free survival (PFS) and overall survival for all patients were 4.1 months and 8.6 months, respectively. Median PFS and overall survival for the platinum-sensitive and platinum-refractory cohorts were 5.4 months versus 3.1 months, and 11.9 months versus 7.1 months, respectively. Toxicity was principally hematologic with grade 3/4 neutropenia in 21% and grade 4 platelets in 8%. There were no treatment-related deaths. Twenty-four of 33 patients (73%) had p53-positive tumors. Although no significant association between platinum sensitivity and p53 status was seen, patients with platinum-sensitive disease and negative p53 by IHC had a trend toward longer survival compared to those with platinum-refractory disease and/or p53 positivity (P = 0.06). We concluded that salvage gemcitabine in this dose and schedule is safe and tolerable in previously platinum-treated patients with NSCLC.