Search

Background aims Bone marrow (BM) is commonly used in the pediatric and adult setting as a source of hematopoietic stem cells (HSCs). The standards of the Joint Accreditation Committee of the International Society for Cell & Gene Therapy & European Society for Blood and Marrow Transplantation (JACIE) include specific requirements regarding BM collection, processing and distribution. To run this process, each transplant team develops a series of JACIE-compliant procedures, customizing them with regard to local settings and paths. Moreover, JACIE standards require that transplant teams validate and periodically revise their procedures to keep the entire process under control. In this article, the authors describe the methodology adopted in our center to fulfill the aforementioned JACIE requirements. Methods The authors developed a validation plan based on the failure mode and effect analysis (FMEA) methodology. According to the FMEA approach, the authors carefully revised activities and procedures connected to BM collection, processing and distribution at our institution. The entire process was initially divided into five main phases (assessment of donor eligibility, perioperative autologous blood donation, preparation of BM collection kit, BM harvesting and BM processing and distribution), comprising 17 subphases and 22 activities. Results For each activity, one or more failure modes were identified, for a total of 28 failure modes, and a risk priority number (RPN) was then assigned to each failure mode. Although many procedures were validated, others were subjected to substantial changes according to the RPN rating. Moreover, specific indicators were identified for subsequent monitoring to contain the risk of failure of steps emerging as critical at FMEA. Conclusions This is the first study describing use of the FMEA methodology within an HSC transplant program. Shaping the risk analysis based on local experience may be a trustworthy tool for identifying critical issues, directing strict monitoring of critical steps or even amending connected procedures. Overall, the FMEA approach enabled the authors to improve our process, checking its consistency over time.

Plerixafor is widely used as up-front treatment with G-CSF to enhance peripheral blood hematopoietic stem cell output in patients failing previous mobilizations. Less frequently, plerixafor is used to rescue an unsatisfactory mobilization following chemotherapy (CT) and G-CSF. This study investigates if pre-collection factors affect the CD34+ cell harvest in chemotherapy and G-CSF mobilizations rescued by plerixafor. Clinical and hematological data relative to patients, mobilization, and apheresis products were retrospectively examined. The outcome was completing a target cell dose ≥ 2 × 106 CD34+ cells/kg at first apheresis. The effect exerted on the outcome by patient- and disease-related factors was investigated by univariate and multivariate logistic regression analysis. The analysis included data from 42 patients affected by hematological (39 patients) and non-hematological malignancies (three patients). Twenty-nine patients (69%) attained the target cell dose at first apheresis. Twelve out of the remaining 13 patients received an additional plerixafor administration, and all accomplished the transplant dose at a second apheresis procedure. Day -1 CD34+ PB count (OR1.46, 95% CI 1.1–1.9, p = 0.008) and platelet count (OR1.0, 95% CI 1.0–1.0, p = 0.033) predicted the achievement of the target dose at first apheresis, independently of pre-mobilization CT, radiation therapy, and disease status at mobilization. At ROC curve analysis, the best cut-off value predicting the successful collection at first apheresis was 7.5/µL for Day -1 CD34+ cell count (AUC 0.830, 0.69 sensitivity, and 0.92 specificity) and 75 × 109/L for Day -1 platelet count (AUC = 0.736, 0.65 sensitivity and 0.85 specificity). In conclusion, on-demand plerixafor rescue allows a successful stem cell collection, irrespectively of disease type and status, prior CT lines, and radiation exposure. Pre-apheresis CD34+ cells and platelet count predict the need for one or two aphereses.

Iron deficiency (ID), with or without anemia, is commonly observed among patients scheduled for cardiac surgery. We investigated if screening ID in the immediate preoperative period and treating ID patients regardless of anemia could reduce perioperative transfusion requirements.This is an observational single-center propensity score-matched study including candidates to elective cardiac surgery prospectively and retrospectively enrolled. Prospectively enrolled patients were screened for ID at hospital admission: if ferritin was ≤100 μg/L or ≤ 300 μg/L with transferrin saturation index ≤20% they received intravenous ferric carboxymaltose, B12-vitamin, and folic acid. A retrospective series of patients not screened for ID and matched for gender, type of surgery, BMI, Goudie transfusion risk score, hemoglobin level, and red blood cell (RBC) indices, served as controls. The primary outcome was the proportion of patients requiring ≤1 packed RBC (pRBC) unit within day 7 or discharge The main secondary outcomes were intraoperative and postoperative pRBC transfusions, duration of hospitalization, and cost-effectiveness of ID screening and treatment.We included 479 prospective and 833 retrospective cases: 442 patients screened for ID and 442 matched controls with unknown iron status were analyzed. ID was observed in 196 patients (44.3%) and iron was administered 1 day (IQR 1-2) before surgery. Overall, 76.9% of patients in the prospective group and 69.7% of controls received ≤1 pRBC transfusion (p = 0.014). The risk for multiple transfusions was lower in patients screened for ID (OR 0.689, 95% CI 0.510-0.930). Despite similar Hb levels at day 7, patients in the prospective group received fewer postoperative pRBC transfusions (p 0.001) and had a shorter hospital length of stay (p 0.001). Globally, hospitalization costs were lower in patients screened and treated for ID.Short-term pre-operative iron therapy is associated with a reduction in postoperative transfusions in anemic and non-anemic ID cardiac surgery patients and has a favorable impact on hospitalization costs.NCT04744181.

Background. Extremely low gestational age neonates (ELGAN, i.e., neonates born before 28 weeks of gestation), are at high risk of developing retinopathy of prematurity (ROP), with potential long-life visual impairment. Due to concomitant anemia, ELGANs need repeated red blood cell (RBC) transfusions. These produce a progressive replacement of fetal hemoglobin (HbF) by adult hemoglobin (HbA). Furthermore, a close association exists between low levels of HbF and severe ROP, suggesting that a perturbation of the HbF-mediated oxygen release may derange retinal angiogenesis and promote ROP. Methods/design. BORN (umBilical blOod to tRansfuse preterm Neonates) is a multicenter double-blinded randomized controlled trial in ELGANs, to assess the effect of allogeneic cord blood RBC transfusions (CB-RBCs) on severe ROP development. Recruitment, consent and randomization take place at 10 Neonatology Intensive Care Units (NICUs) of 8 Italian tertiary Hospitals. ELGANs with gestational age at birth comprised between 24 + 0 and 27 + 6 weeks are randomly allocated into two groups: 1) standard RBC transfusions (adult-RBCs) (control arm); 2) CB-RBCs (intervention arm). In case of transfusion need, enrolled patients receive transfusions according to the allocation arm, unless an ABO/RhD CB-RBC is unavailable. Nine Italian public CB banks cooperate to make available a suitable amount of CB-RBC units for all participating NICUs. The primary outcome is the incidence of severe ROP (stage 3 or higher) at discharge or 40 weeks of post-menstrual age, which occurs first. Discussion. BORN is a ground-breaking trial, pioneering a new transfusion approach dedicated to ELGANs at high risk for severe ROP. In previous non-randomized trials, this transfusion approach was proven feasible and able to prevent the HbF decrease in patients requiring multiple transfusions. Should the BORN trial confirm the efficacy of CB-RBCs in reducing ROP severity, this transfusion strategy would became the preferential blood product to be used in severely preterm neonates. Trial registration: ClinicalTrials.gov Identifier: NCT05100212

Background: The coronavirus disease 2019 (COVID-19) pandemic has deeply modified the complex logistical process underlying allogeneic hematopoietic stem cell transplant practices. Aim: In light of these changes, the authors compared data relative to allogeneic transplants carried out from 2018 at their center before (n = 167) and during the pandemic (n = 45). Methods: The authors examined patient characteristics, donor and graft types, cell doses and main transplant outcomes. Moreover, the authors evaluated the rise of costs attributable to additional COVID-19-related procedures as well as the risk of adverse events these procedures conveyed to grafts or recipients. Results: Overall, the number of transplants did not decrease during the pandemic, whereas patients at high relapse risk were prioritized. Transplants were mainly from matched unrelated donors, with a significant decrease in haploidentical related donors. Moreover, the use of bone marrow as a graft for haploidentical transplant was almost abandoned. Cryopreservation was introduced for all related and unrelated apheresis products, with a median storage time of 20 days. Notably, transplant outcomes (engraftment, acute graft-versus-host disease and non-relapse mortality) with cryopreserved products were comparable to those with fresh products. Conclusions: Considering that the emergency situation may persist for months, cryopreserving allogeneic grafts can offer a lifesaving opportunity for patients whose allogeneic transplant cannot be postponed until after the end of the COVID-19 pandemic.

Background: Hypoxia-inducible factor (HIF-1) is a transcription factor produced in hypoxia condition, it is closely associated with tumor angiogenesis and metastasis. Aim: To investigate the expression of HIF-1α in relation with the presence or absence of bone metastasis. Methods A retrospective analysis was carried out on samples deriving from bronchial biopsy and CT-guided trans-thoracic needle biopsy. Detection of HIF-1 expression was performed on tissue sample by a monoclonal murine antibody, comparing patients with or without bone metastases (BM+). Findings: In the total population the main histotype was adenocarcinoma (71.5%), COPD the prevalent comorbidity (73.6%), the mean pack-year was 36.4. Ninety-five histology samples were considered for analysis and comparison. Subdividing the population according to the presence or not of bone metastases, significant differences were found in pack-years (p = 0.02), time to progression (TTP) (p = 0.001) and COPD comorbidity (p = 0.04). The survival comparison between the two subgroups obtained by Kaplan–Meier method showed a longer TTP in patients with visceral metastases with a HR of 1.3 though the comparison by this method was not significant (p = 0.1). A higher intensity and percentage of expression of HIF-1α was recorded in the group with bone metastases (p = 0.02). The main variable affecting HIF expression in a multivariate analysis was the presence of bone metastases (p = 0.01). Interpretation: Patients affected by NSCLC IV stage with bone metastasis have lower survival. There is a very close link between bone metastasis and HIF-1α expression level. The latter could be considered a predictive factor of bone spread and poor prognosis.

Repeated red blood cell (RBC) transfusions are thought to increase the risk for retinopathy of prematurity (ROP), likely due to a critical fetal hemoglobin (HbF) reduction. In this study, we investigated if the postmenstrual age (PMA) of neonates at transfusion influences the risk for ROP. We estimated the cumulative transfusion-free survival (TFS) in a series of 100 preterm neonates receiving one or more RBC units. TFS was calculated by censoring patients at first transfusion and expressing the time between birth and transfusion as either PMA or postnatal day. Then, we investigated if TFS predicted the occurrence of severe ROP, defined as ROP stage 3 or higher. We found that neonates with severe ROP displayed a significantly shorter TFS expressed according to their PMA (p = 0.001), with similar TFS according to postnatal days. At receiver operating characteristic (ROC) curve analysis, receiving an RBC unit before week 28 of PMA predicted severe ROP with a sensitivity of 64% and a specificity of 78%. In addition, receiving a second RBC unit before the PMA of 29 weeks predicted severe ROP with a sensitivity of 75% and a specificity of 69%. At multivariate analysis, PMA at the second transfusion was even more informative than at first transfusion and outperformed all other variables in predicting severe ROP, with an odds ratio of 4.554 (95% CI 1.332–15.573, p = 0.016). Since HbF decrease is greater after multiple RBC transfusions, it is conceivable that neonates receiving more than one unit before the PMA of 29 weeks may be exposed to a greater disturbance of retinal vascularization. Any strategy aimed at preventing the critical HbF decrease at this low age might potentially reduce the risk for severe ROP.

Supplemental Digital Content is available in the text., Background. The impact of ABO incompatibility (ABO-I) on hematopoietic stem cell transplant outcomes is still debated. Methods. We retrospectively investigated 432 consecutive transplants performed at our center (2012–2020). All patients but 6 were affected by hematologic malignancies. The effect of different ABO match combinations on engraftment rate, transfusion support, acute and chronic graft-versus-host disease incidences, nonrelapse mortality (NRM), disease-free survival, and overall survival was assessed in univariate and multivariate analysis. Significance was set at P

Objectives The primary objective is to demonstrate that COVID-19 convalescent plasma (CCP) prevents progression to severe pneumonia in elderly COVID-19 pneumonia patients with chronic comorbidities. Secondary objectives are to demonstrate that CCP decreases the viral load in nasopharyngeal swabs and increases the anti-SARS-CoV-2 antibody titre in recipients. Trial design This is a randomized, open-label, parallel group, phase II/III study with a superiority framework. The trial starts with a screening phase II designed with two-tailed alpha=0.2. In case of positive results, the trial will proceed in a formally comparative phase III (alpha=0.05). Participants Adult patients with confirmed or suspected COVID-19 who are at risk according to CDC definition are eligible. Inclusion criteria are all the following: age ≥ 65; pneumonia at CT scan; PaO2/FiO2 ≥300 mmHg; presence of one or more comorbidities; signed informed consent. Exclusion criteria are one of the following: age < 65; PaO2/FiO2 < 300 mmHg; pending cardiopulmonary arrest; refusal to blood product transfusions; severe IgA deficiency; any life-threatening comorbidity or any other medical condition which, in the opinion of the investigator, makes the patient unsuitable for inclusion. The trial is being conducted at three reference COVID-19 centres in the middle of Italy. Intervention and comparator Intervention: COVID-19 Convalescent Plasma (CCP) in addition to standard therapy. Patients receive three doses (200 ml/day on 3 consecutive days) of ABO matched CCP. Comparator: Standard therapy Main outcomes A. Primary outcome for Phase II: Proportion of patients without progression in severity of pulmonary disease, defined as worsening of 2 points in the ordinal scale of WHO by day 14. B. Primary outcome for Phase III: Proportion of patients without progression in severity of pulmonary disease, defined as worsening of 2 points in the ordinal scale of WHO by day 14. Secondary outcomes for Phase III: Decreased viral load on nasopharyngeal swab at days 6, 9 and 14; Decreased viremia at days 6 and 9; Increased antibody titer against SARS-CoV2 at days 30 and 60; Proportion of patients with negative of SARS-CoV2 nasopharyngeal swab at day 30; Length of hospital stay; Mortality rate at day 28; Total plasma related adverse event (day 60); Total non-plasma related adverse events (day 60); Severe adverse events (SAE) (day 60). Randomisation Treatment allocation is randomized with a ratio 1:1 in both phase II and phase III. Randomization sequences will be generated at Fondazione Policlinico Gemelli IRCCS through the RedCap web application. Randomized stratification is performed according to age (under/over 80 years), and sex. Blinding (masking) None, this is an open-label trial. Numbers to be randomised (sample size) Phase II: 114 patients (57 per arm). Phase III: 182 patients (91 per arm) Trial Status The trial recruitment started on May 27, 2020. The anticipated date of recruitment completion is April 30, 2021. The protocol version is 2 (May 10, 2020). Trial registration The trial has been registered on ClinicalTrials.gov (May 5, 2020). The Identifier number is NCT04374526 Full protocol The full protocol is attached as an additional file, accessible from the Trials website (Additional file 1). In the interest in expediting dissemination of this material, the familiar formatting has been eliminated; this Letter serves as a summary of the key elements of the full protocol.

Neonates and prematures are among the most transfused categories of patients. Adverse reactions due to transfusions, such as transfusion-transmitted infections, can affect the rest of their lives. In this systematic review, we revised the literature concerning transfusion-transmitted infection in neonates. We reported case-reports and case-series previously published and we integrated these data with our experience at local neonatal intensive care unit. Moreover, we illustrated strategies for mitigating transfusion-transmitted infections, including donor selection and testing, pathogen inactivation technologies and combined approaches, as for Cytomegalovirus infection, integrating leukoreduction and identification of seronegative donors.

Repeated red blood cell (RBC) transfusions in preterm neonates are associated with poor outcome and increased risk for prematurity-associated diseases. RBC transfusions cause the progressive replacement of fetal haemoglobin (HbF) by adult haemoglobin (HbA). We monitored HbF levels in 25 preterm neonates until 36 weeks of post-menstrual age (PMA); patients received RBC units from allogeneic cord blood (cord-RBCs) or from adult donors (adult-RBCs), depending on whether cord-RBCs were available. Primary outcome was HbF level at PMA of 32 weeks. Twenty-three neonates survived until this age: 14 received no transfusions, two only cord-RBCs, three only adult-RBCs and four both RBC types. HbF levels in neonates transfused with cord-RBCs were significantly higher than in neonates receiving adult-RBCs (P < 0·0001) or both RBC types (P < 0·0001). Superimposable results were obtained at PMA of 36 weeks. Every adult-RBCs transfusion increased the risk for an HbF in the lowest quartile by about 10-fold, whereas this effect was not evident if combined adult- and cord-RBCs were evaluated. Overall, these data show that transfusing cord-RBCs can limit the HbF depletion caused by conventional RBC transfusions. Transfusing cord blood warrants investigation in randomised trials as a strategy to mitigate the severity of retinopathy of prematurity (NCT03764813).

We analyzed data relative to cell content in 88 consecutive patients receiving HLA haploidentical bone marrow (BM) transplants with post-transplantation cyclophosphamide (PT-CY). Median age was 54.5 (range, 17–72); diagnoses were acute leukemia (n = 46), lymphoproliferative disorders (n = 24), myelofibrosis (n = 11) and myelodysplastic syndromes (n = 5). Total nucleated cell (TNC) and CD34+, CD3+, CD4+ and CD8+ cell doses were stratified as higher than first, second and third quartile and the dose effect on various clinical outcomes was assessed. Median time to engraftment was 17 days for neutrophils and 24 days for platelets. To receive a dose of TNC ≥3.2 x 106/kg or CD34+ cells ≥2.7 x 106/kg significantly shortened the time to neutrophil and platelet engraftment and reduced the blood product requirements in the 30-day period after transplantation. Overall, TNC and CD34+ cell doses had no effect on acute graft-versus-host disease (GVHD) incidence, whereas patients receiving higher CD3+ and CD8+ cell doses seemed to have less chronic GVHD. No effect on non-relapse mortality, progression-free survival and overall survival was observed at different cell dose thresholds. These data suggest that in HLA haploidentical BM transplant with PT-CY, appropriate cell doses are relevant to the engraftment. The association between low CD3+/CD8+ cells and chronic GVHD deserves further investigation.

Platelet-rich plasma (PRP) is an inexpensive and safe substitute of recombinant growth factors in Vitro and in vivo. Due to its putative effect on tissue repair, the use of autologous PRP has become largely popular. Recently, a jellified PRP derivative obtained from umbilical cord blood (CB) has been utilized in vivo to treat mucosal and cutaneous lesions. Nevertheless, whether PRP derived from CB and adult blood display different potency in promoting cell growth in Vitro has been rarely investigated. In this study, we compared cytokine profile and mesenchymal cell growth supporting the ability of platelet lysate obtained from adult and cord blood. Our in vitro results strongly back the utilization of CB platelet lysate in vivo, as an efficacious, safe and inexpensive alternative to promote damaged tissue healing when the autologous PRP is contraindicated. Moreover, the policy of manufacturing CB platelet lysate can limit the current disposal of many collected CB units not suitable for transplant due to their low nucleated cell count.

Background Human cord blood (CB) endothelial colony forming cells (ECFCs) are endowed with high vascular regenerative ability in immunodeficient mice, but their immunogenicity and susceptibility to rejection in immunocompetent models has yet to be explored. Methods We injected CB ECFCs in non-immuno-suppressed C57BL/6 J mice after having induced the hindlimb ischemia and we investigated their contribution to the recovery from the ischemic injury. Human ECFCs (hECFCs) were administered by intramuscular injection and hindlimb blood perfusion was measured by laser Doppler analysis at 7-day intervals for 28 days after treatment. Mice were sacrificed after 7 and 28 days and immunohistochemistry for specific human (CD31) and mouse (von Willebrand factor) endothelial antigens was carried out. Before euthanasia, blood samples to assess cytokines and angiogenic growth factor levels were collected. Results Mice injected with hECFCs showed a prompter and greater recovery of blood flow than controls. Several endothelial cells of human origin were detected at day7 after injection and their number declined progressively. Likewise, a progressive increase of mouse-derived vascular structures were observed, paralleled by the amplified endogenous production of various soluble mediators of angiogenesis, including Vascular Endothelial Growth Factor and Fibroblast Growth Factor. Conclusions Overall, our findings are consistent with the hypothesis that human ECFCs might expand the endogenous vascular repair potential of recipients and support their possible HLA-independent unconventional use.

The past 20 years of experience with umbilical cord blood transplantation have demonstrated that cord blood is effective in the treatment of a spectrum of diseases, including hematological malignancies, bone marrow failure, hemoglobinopathies, and inborn errors of metabolism. However, only a few number of umbilical cord blood units collected have a cell content adequate for an allogenic hematopoietic stem cell transplantation. In the meanwhile, there is an increasing interest in exploiting cord blood derivatives in different fields. In this review, we will summarize the most recent updates on clinical applications of umbilical cord blood platelet derivatives for regenerative medicine, and we will revise the literature concerning the use of umbilical cord blood for autologous or allogeneic transfusion purposes. The methodological aspect and the biological characteristics of these products also will be discussed.

The weak D phenotype is represented by a group of RHD genotypes that code for alterated RhD proteins associated with a reduced RhD expression on red blood cell. By routine serology, some partial D variants are likely to be missed. In this report we describe the case of a three-year-old Black African child with a “unclear” reaction with monoclonal anti-D. We analyzed the blood sample of the child with different methods to conclude that it is a case of DAR 1.2 (weak D 4.2.2) and that it must be transfused with D negative erithrocytes.

Chronic red blood cell transfusions remain an essential part of supportive treatment in patients with thalassaemia and sickle cell disease (SCD). Red blood cell (RBC) transfusions expose patients to the risk of developing antibodies: RBC alloimmunization occurs when the immune system meets foreign antigens. We created a register of extensively genotyped donors to achieve a better matched transfusion in order to reduce transfusion alloimmunization. Extended RBC antigen typing was determined and confirmed by molecular biology techniques using Human Erythrocyte Antigen (HEA) BeadChip (BioArray Solutions Ltd., Warren, NJ) in periodic blood donors and in patients with thalassaemia and SCD. During 3 years, we typed extensively 1220 periodic blood donors, 898 male and 322 female. We also studied 10 hematologic patients affected by thalassaemia and sickle cell disease referred to our institution as candidate to periodic transfusions. Our patients (8 females and 2 males with a median age of 48 years, range 24-76 years), extensively typed using molecular techniques and screened for RBC alloantibodies, were transfused with a median of 33.5 RBC units. After three years of molecular typing, the "perfect match" transfusion strategy avoided new alloantibodies development in all studied patients.

It is still under debate whether granulocyte transfusions (GTs) substantially increase survival in patients with febrile neutropenia. We retrospectively examined data relative to 96 patients with hematological malignancies receiving 491 GTs during 114 infectious episodes (IE). Patients were grouped according to the median doses of granulocytes transfused during the infectious episode (low-dose group: 3.0x108 cells/Kg). The impact of clinical, microbiological and GT-related variables on the infection-related mortality (IRM) was investigated. The IRM was not influenced by the number of GTs or by the total amount of granulocytes received, whereas a dose-related effect of the median dose received for IE was detected at univariate analysis (IRM of 18.4% in the standard-dose group, 44.4% in the low-dose group and 48.4% in the high-dose group, p = 0.040) and confirmed at multivariate analysis (OR 3.7, IC 95% 1.5-8.9; 0.004 for patients not receiving standard doses of GTs). Moreover, patients receiving GTs at doses lower or greater than standard had increased risk for subsequent ICU admission and reduced overall survival. The dose-related effect of GTs was confirmed in bacterial but not in fungal infections. Preliminary findings obtained from a subgroup of patients candidate to GTs revealed that levels of inflammatory response mediators increase in a dose-related manner after GTs, providing a possible explanation for the detrimental effect exerted by high-dose transfusions. GTs can constitute a valuable tool to improve the outcome of infections in neutropenic patients, provided that adequate recipient-tailored doses are supplied. Further investigations of the immunomodulatory effects of GTs are recommended.

Purpose This phase III, randomized, open-label, multicenter study compared the overall survival associated with irinotecan plus gemcitabine (IRINOGEM) versus gemcitabine monotherapy (GEM) in patients with chemotherapy-naive, locally advanced or metastatic pancreatic cancer. Patients and Methods IRINOGEM patients received starting doses of gemcitabine 1,000 mg/m2 and irinotecan 100 mg/m2 given weekly for 2 weeks every 3-week cycle. GEM patients received gemcitabine 1,000 mg/m2 weekly for 7 of 8 weeks (induction) and then weekly for 3 of 4 weeks. The primary end point of the trial was survival. Secondary end points included tumor response, time to tumor progression (TTP), changes in CA 19-9, and safety. Results In each arm, 180 randomly assigned patients comprised the intent-to-treat population evaluated for efficacy; 173 IRINOGEM and 169 GEM patients were treated. Median survival times were 6.3 months for IRINOGEM (95% CI, 4.7 to 7.5 months) and 6.6 months for GEM (95% CI, 5.2 to 7.8 months; log-rank P = .789). Tumor response rates were 16.1% (95% CI, 11.1% to 22.3%) for IRINOGEM and 4.4% (95% CI, 1.9% to 8.6%) for GEM (χ2 P < .001). Median TTP was 3.5 months for IRINOGEM versus 3.0 months for GEM (log-rank P = .352). However, subset analyses in patients with locally advanced disease suggested a TTP advantage with IRINOGEM versus GEM (median, 7.7 v 3.9 months). CA 19-9 progression was positively correlated with tumor progression. The incidence of grade 3 diarrhea was higher in the IRINOGEM group but grade 3 to 4 hematologic toxicities and quality-of-life outcomes were similar. Conclusion IRINOGEM safely improved the tumor response rate compared with GEM but did not alter overall survival.