The role of human endogenous retroviruses in melanoma

Sequencing of the human genome has established that our DNA harbours many endogenous retrovirus (ERV) sequences, remnants of ancestral exogenous retroviral infections fixed in the germline DNA. In recent years, human ERVs (HERVs) have been implicated in melanomagenesis. Retrovirus-like particles and the expression of HERV mRNA and proteins have been demonstrated in melanoma tissue. In addition, antibodies to HERV proteins have been observed in patients with melanoma. In vitro and mouse models have provided fascinating insights into the potential mechanisms of HERVs in melanomagenesis. This review considers the evidence associating HERVs with melanoma.

Anti-social personality disorder and response to methadone maintenance treatment

A sample of 183 current methadone maintenance patients were interviewed on their drug use history, criminal history, current drug use, and symptoms of Anti-social Personality Disorder (ASPD). Thirty-nine percent of patients met the DSM-III-R criteria for a diagnosis of ASPD. ASPD patients had an earlier onset of drug use, drug injecting, heroin use, had wider polydrug using histories and had been arrested earlier and more frequently than other patients. Despite the different pretreatment histories of ASPD and other patients, there were no differences between the two groups in retention in treatment, methadone dosage or heroin use. It is concluded that heroin-dependent ASPD patients can be successfully retained in methadone treatment, on similar methadone doses and with similar in-treatment drug use patterns as those of non-ASPD heroin dependent patients.

Health-related quality of life/patient-reported outcomes in relapsed ovarian cancer: Results from a randomized phase III study of trabectedin with pegylated liposomal doxorubicin (PLD) versus PLD alone

5526

Background: In an open-label, multicenter, randomized phase III study comparing the combination of trabectedin and PLD to PLD alone in patients with relapsed ovarian cancer, the combination demonstrated significantly improved progression free survival and response rates, manageable non-cumulative toxicity, and fewer PLD-associated adverse events. We studied the impact of the combination of trabectedin with PLD on the quality of life (QoL)/patient-reported outcomes (PRO) evaluated as part of the trial. Methods: QoL/PRO questionnaires, EORTC-QLQ C30, OV28, and EQ-5D were completed by patients at screening and on Day 1 of every other treatment cycle starting with Cycle 1, and at the end-of-treatment visit. Global health status/QoL, fatigue, rain subscales from QLQ C30, and abdominal pain/GI symptoms scale from OV28 were chosen a priori for primary analyses. Other scales of the three questionnaires were analyzed on a supportive basis. Results: A total of 672 patients were randomized. 663 (98%) completed at least the baseline questionnaires. Median cycles of treatment was 6 (131 days) for the combination arm and 5 (143 days) for the monotherapy arm. Mixed effects models (using a covariance structure of AR[1]) predicting the score at baseline and follow-up scores as a function of treatment, days after baseline, and interaction between treatment and days after baseline showed no significant differences between the treatment arms for any of the prespecified scales. Similar analyses of other scales, including EQ-5D Health Index scores and Health State on the Visual Analog Scale, support the findings. Conclusions: The addition of trabectedin to PLD results in superior efficacy in patients with relapsed ovarian cancer, with no added decrement to overall health status as assessed by PRO.

[Table: see text]

Randomized phase II study of decitabine in combination with carboplatin compared with carboplatin alone in patients with recurrent advanced ovarian cancer

5562

Background: Experimental and clinical data (Gifford et al, Clin Cancer Res. 2004) indicate the potential importance of methylation in mediating resistance to carboplatin in ovarian cancer. A previous phase I trial (Appleton et al, J Clin Oncol. 2007) established the feasibility of combining carboplatin with the demethylating agent decitabine on a day 1 + 8 q4 weekly (w) schedule and PD data provided evidence of target cell demethylation. Methods: Patients (pt) with ovarian cancer relapsing 6–12 months following first line treatment were randomised to receive either 6 cycles of carboplatin AUC 6 q4 w (Arm A), or 90 mg/m2 decitabine as a 6 hour infusion on day 1 and carboplatin AUC 6 on day 8 q4 w (Arm B). The primary endpoint was response rate. An interim analysis was planned after 11 patients were enrolled into Arm B. Results: 29 pt were enrolled. After the first 4 pt had been treated (at 90 mg/m2 decitabine) the frequency of dose delays due to neutropenia was considered unacceptable, and therefore the starting dose of decitabine was reduced to 45 mg/m2 for the subsequent 11 pt. 7 out of 14 pt in Arm A completed 6 cycles compared with 0 of 11 in Arm B (at 45 mg/m2 decitabine). Grade 2/3 hypersensitivity reactions were more common in Arm B than Arm A (64% vs. 21%), as were prolonged treatment delays due to neutropenia (36% vs. 10%). At the interim analysis, in the 11 pt treated with 45mg/m2 (Arm B), there were no RECIST responses, while 2 pt had short-lived CA125 responses (59 and 63 days). In contrast 6 of 14 patients in Arm A had RECIST responses consistent with the expected efficacy of carboplatin in this population. Conclusions: The lack of efficacy, as well as the difficulties in treatment delivery in Arm B, led the project team to conclude that the study should be closed. With this dose and schedule, decitabine is ineffective in reversing carboplatin resistance. Further investigations are ongoing to understand (a) the apparent increased incidence of hypersensitivity and (b) the trend towards reduced efficacy in Arm B.

[Table: see text]

Phase I feasibility study of carboplatin plus capecitabine followed by maintenance capecitabine in patients (pts) with recurrent platinum-sensitive epithelial ovarian cancer (EOC)

5564

Background: In a previous study, we noted a response rate (RR) of 61% for the 3 drug combination of carboplatin, epirubicin and capecitabine in platinum-sensitive recurrent EOC. This combination however resulted in excessive grade (G) 3–4 haematological toxicity (55%) (BJC 2006; 94:74). The current trial therefore assessed the feasibility and efficacy of the 2 drugs, carboplatin and capecitabine as second- or third-line treatment. Methods: Pts were administered carboplatin (AUC5) day 1 and capecitabine at a starting dose of 750 mg/m2 bd, days 1–21, q21 (dose level 1). The capecitabine dose was deescalated to 625 mg/m2 (dose level -1) and 500 mg/m2 (dose level -2) according to toxicity. Pts with an objective response or stable disease received maintenance capecitabine (at the same dose level) for up to 12 months or until progression. Responses were assessed with RECIST criteria and CA-125. Results: 19 of the 20 pts enrolled were evaluable for toxicity and response. Dose-limiting toxicity was observed at dose level 1 (G3 fatigue, G3 diarrhoea, G3 neutropenia of > 14 days; n = 3/5), dose level -1 (G3 angina (n = 2), G3 vomiting, G3 palmar plantar erythema; n = 4/7) and dose level-2 (diarrhoea / fatigue; n = 1/7). One patient had a G3 carboplatin hypersensivity reaction. 8 pts received maintenance capecitabine which was well tolerated. The overall RR was 53% with 10 partial responses and 5 stable diseases. The median progression free survival (PFS) was 6.5 months (m) and the 6mPFS was 63% with 2 pts currently ongoing treatment. The median PFS on maintenance was 3.2 m. Conclusions: The combination was well tolerated at the recommended phase II dose of carboplatin (AUC 5) and capecitabine (500 mg/m2 bd) with partial responses in over half of the cases.

[Table: see text]

Correlation between electron transport and shear Alfvén activity in the National Spherical Torus Experiment

We report the observation of a correlation between shear Alfvén eigenmode activity and electron transport in plasma regimes where the electron temperature gradient is flat, and thus the drive for temperature gradient microinstabilities is absent. Plasmas having rapid central electron transport show intense, broadband global Alfvén eigenmode (GAE) activity in the 0.5-1.1 MHz range, while plasmas with low transport are essentially GAE-free. The first theoretical assessment of a GAE-electron transport connection indicates that overlapping modes can resonantly couple to the bulk thermal electrons and induce their stochastic diffusion.

HLA-associated clinical progression correlates with epitope reversion rates in early human immunodeficiency virus infection

Human immunodeficiency virus type 1 (HIV-1) can evade immunity shortly after transmission to a new host but the clinical significance of this early viral adaptation in HIV infection is not clear. We present an analysis of sequence variation from a longitudinal cohort study of HIV adaptation in 189 acute seroconverters followed for up to 3 years. We measured the rates of variation within well-defined epitopes to determine associations with the HLA-linked hazard of disease progression. We found early reversion across both the gag and pol genes, with a 10-fold faster rate of escape in gag (2.2 versus 0.27 forward mutations/1,000 amino acid sites). For most epitopes (23/34), variation in the HLA-matched and HLA-unmatched controls was similar. For a minority of epitopes (8/34, and generally associated with HLA class I alleles that confer clinical benefit), new variants appeared early and consistently over the first 3 years of infection. Reversion occurred early at a rate which was HLA-dependent and correlated with the HLA class 1-associated relative hazard of disease progression and death (P = 0.0008), reinforcing the association between strong cytotoxic T-lymphocyte responses, viral fitness, and disease status. These data provide a comprehensive overview of viral adaptation in the first 3 years of infection. Our findings of HLA-dependent reversion suggest that costs are borne by some escape variants which may benefit the host, a finding contrary to a simple immune evasion paradigm. These epitopes, which are both strongly and frequently recognized, and for which escape involves a high cost to the virus, have the potential to optimize vaccine design.

Neurological morbidity among human T-lymphotropic-virus-type-1-infected individuals in a rural West African population

BACKGROUND

Community-based neurological data about human T lymphotropic virus type 1 (HTLV-1) morbidity in sub-Saharan Africa are scarce.

OBJECTIVES

To ascertain the prevalence of neurological morbidity, in particular tropical spastic paraparesis (TSP), among HTLV-1-infected subjects and to compare TSP prevalence in HTLV-1-infected with that in non-infected subjects in a rural West African population.

METHODS

A cross-sectional study of HTLV-1-infected cases and controls (ratio 4:1) from a rural community (population approximately 10 000, HTLV-1 prevalence 7.7%). One neurologist masked to HTLV-1 serological status assessed all subjects. Clinical criteria were employed to diagnose TSP.

RESULTS

From 205 eligible cases and controls, 139 were recruited with a mean age of 56 years, and 113 (81%) were HTLV-1-infected. 108/139 (78%) were female, and 8/113 HTLV-1 infected cases (7.1%) had a definite or probable TSP (all females; mean age 67 years) compared with 0/26 controls. Two with TSP were co-infected with HIV-2. Complaints of back pain and leg weakness were more common in HTLV-1-infected individuals (p = 0.03, p = 0.02), but no single symptom distinguished between subjects with and without TSP.

CONCLUSION

We report a prevalence of TSP among HTLV-1-infected persons in this rural West African setting of 7.1%. There are difficulties excluding other potential aetiologies here.

A Phase II study of trabectedin single agent in patients with recurrent ovarian cancer previously treated with platinum-based regimens

The objective of this study was to determine the objective response rate in patients with platinum-sensitive and platinum-resistant recurrent ovarian cancer to treatment with trabectedin (Yondelis®) administered as a 3-h infusion weekly for 3 weeks of a 4-week cycle. We carried out a multicentre Phase II trial of trabectedin in patients with advanced recurrent ovarian cancer. Trabectedin (0.58 mg m⁻²) was administered via a central line, after premedication with dexamethasone, to 147 patients as a 3-h infusion weekly for 3 weeks followed by 1-week rest. Major eligibility criteria included measurable relapsed advanced ovarian cancer and not more than two prior platinum-containing regimens. Patients were stratified according to the treatment-free interval (TFI) between having either platinum-sensitive (⩾6 months TFI) or platinum-resistant disease (<6 months TFI)/platinum-refractory disease (progression during first line therapy). In the platinum-sensitive cohort, 62 evaluable patients with measurable disease had an overall response rate (ORR) of 29.0% (95% CI: 18.2–41.9%) and median progression-free survival (PFS) was 5.1 months (95% CI: 2.8–6.2). Four patients with measurable disease per Response Evaluation Criteria in Solid Tumours (RECIST) criteria had no follow-up scans at the end of treatment. In the platinum-resistant/refractory cohort, 79 patients were evaluable with an ORR of 6.3% (95% CI: 2.1–14.2%). Median PFS was 2.0 months (95% CI: 1.7–3.5 months). Two patients with measurable disease per RECIST criteria had no follow-up scans at the end of treatment. The most frequent (⩾2% of patients) drug-related treatment-emergent grade 3/4 adverse events were reversible liver alanine transferase elevation (10%), neutropaenia (8%), nausea, vomiting, and fatigue (5% each). Trabectedin is an active treatment, with documented responses in patients with platinum sensitive advanced relapsed ovarian cancer, and has a manageable toxicity profile.

Microtearing instabilities and electron transport in the NSTX spherical tokamak

We report a successful quantitative account of the experimentally determined electron thermal conductivity chi(e) in a beam-heated H mode plasma by the magnetic fluctuations from microtearing instabilities. The calculated chi(e) based on existing nonlinear theory agrees with the result from transport analysis of the experimental data. Without using any adjustable parameter, the good agreement spans the entire region where there is a steep electron temperature gradient to drive the instability.

Final results of a phase I study of wild-type oncolytic reovirus administered intravenously to patients with advanced cancer

3572

Background: Reovirus has minimal pathogenicity in humans but selectively replicates in cells with activated Ras, which inactivates the anti-viral effects of double stranded RNA-activated protein kinase. Wild-type reovirus serotype 3 Dearing has selective antitumor activity in pre-clinical models. Methods: Wild-type reovirus serotype 3 Dearing was administered as a 1-hr IV infusion every 4 weeks, initially for 1 day, then for 3 and then 5 days every 4 weeks. The starting dose was 1 x 10^8 tissue culture infectious dose (TCID50), increasing in successive cohorts until the observation of grade 2 toxicity. Endpoints were safety, detection of viral replication, viral shedding, immune response and antitumor activity. Results: A total of 33 patients (25 males, median age 61; ECOG performance status 0/1/2 = 15/17/1) were treated. 8 cohorts were treated at dose levels of 1x10^8 TCID50 for 1 day, 1x10^8 for 3 days, and 1x10^8, 3x10^8, 1x10^9, 3x10^9, 1x10^10 and 3x10^10 for 5 days. Maximum tolerated dose (MTD) was not reached and toxicities were mild (grade 1 or 2), including chills, fever, headache, rhinorrhea, fatigue and myelosuppression. Reverse transcription polymerase chain reaction (RT-PCR) studies of blood, urine, stool and sputum post reovirus administration and every week thereafter were negative for viral shedding for all treated patients. Anti-reovirus antibodies were present at baseline in most patients, and titres increased after cycle 1 by between 1- and 4-log. Intratumoral reovirus replication was detected by electron microscopy in tumor biopsies. PSA decreased by 50% in a patient with metastatic prostate cancer treated at 3x10^9 TCID50, with evidence of tumor necrosis on CT scanning. 2 patients with metastatic colorectal cancer treated at 3x10^8 and 3x10^9 TCID50 had CEA reductions of 60% and 27%, and received 6 and 3 cycles respectively. 1 patient with metastatic bladder cancer treated at 1x10^9 TCID50 had a minor response and received 4 cycles. Conclusions: Wild-type reovirus serotype 3 Dearing is well tolerated, with minimal toxicity, and MTD has not been reached. Despite a vigorous host response to virus, tumor necrosis associated with intratumoral viral replication was observed. Clinical indications of anticancer activity were seen and phase II studies are planned.

[Table: see text]

A phase II study of the feasibility of sequential carboplatin (C) followed by paclitaxel (P) and gemcitabine (G) as first-line chemotherapy for stage Ic-IV ovarian (OC), fallopian tube (FTC), and primary peritoneal carcinoma (PPC)

5578

Background: We previously explored the feasibility of adding G to C + P in a phase II study, in which 4 cycles of C followed by 4 cycles of concurrent weekly P and G were given (BJC (2004) 91:627–32). Although highly active (med PFS 19.5m), the feasibility of this 1st-line regimen was limited by dyspnea during the weekly P+G phase. The current trial assessed whether the dyspnoea could be ameliorated by altering the schedule. Methods: Untreated FIGO stage Ic-IV OC, FTC and PPC patients (pts) were eligible. Chemotherapy (CTX) consisted of 4 cycles of C (AUC 7) q21 days, followed by 4 cycles of concurrent P (175 mg/m2) d1 and G (1,000 mg/m2) d1 and d8 q21 days. The primary endpoint of the study was the percentage of pts completing the planned 8 cycles of CTX. The planned sample size was 54 pts, based on a one stage single arm study design with 95% power to reject the null hypothesis (completion rate less than 60%), assuming a true completion rate of 80% using one-sided alpha=0.05. Results: All 54 pts were recruited between June 05 and June 06. Details for 44 pts are currently available for the C phase. 38% of these pts had one or more dose reduction (DR), and 68% had dose delays (DD). The commonest reason for DR and DD was neutropenia. Details for 33 pts are currently available for the P+G phase. 27% of these pts had a DR of P and 51% had a DR of G. 71% of pts omitted G on D8, and 64% had a DD. The commonest reasons for DR and DD were neutropenia and thrombocytopenia. Overall, the KM estimate of the percentage completing 8 cycles is 75% (95% CI 61%-89%) based on the current data. Dyspnoea (Grade 2 only) was observed in 4.5% and 3% pts during the C and P+G phases respectively. No significant treatment related CXR changes were observed. The overall response rate was 68% (95% CI 45–86%; 36% CR, 32%PR, 14% SD, 5%PD, 14% NE; n=22). The median follow-up is 7.5 months and the PFS at 8 months is 92% (95% CI 77–97%). Conclusion: This schedule appears to ameliorate the previously observed dyspnoea, while retaining comparable efficacy. The feasibility of this regimen is limited by myelosuppression which could potentially be overcome by: a) reducing the dose of gemcitabine to 750 mg/m2, or b) using prophylactic Peg-GCSF.

[Table: see text]

Single agent, canfosfamide (C, TLK286) vs pegylated liposomal doxorubicin (D) or topotecan (T) in 3rd-line treatment of platinum (P) refractory or resistant ovarian cancer (OC): Phase 3 study results

LBA5528

Background: Canfosfamide (C) is a novel glutathione analog prodrug activated by glutathione S-transferase P1–1. P resistant OC has a poor prognosis and non-P agents are used for palliation. C reported objective responses in multicenter Phase 2 OC trials. Methods: Pts with P resistant OC who progressed after D or T, measurable disease (RECIST), adequate liver/renal/bone marrow function were eligible. Randomization was stratified by prior D or T treatment, ECOG PS (0 or 1 vs 2) and presence or absence of bulky disease (= 5cm). Patients received C at 1000 mg/m2 IV q3wks, or to D at 50 mg/m2 IV q4wks or T at 1.5 mg/m2 IV daily × 5 q3wks until progression. The trial had a 90% power to detect a 29% reduction in the relative risk of death. Results: 461 pts (C=232 and D or T=229) received 1052 cycles [median 3; range (r) 1–33], 699 (median 4; r 1–32), and 469 (median 5; r 1–21) for C, D and T respectively. Most common Grade 3–4 AEs for C were: nausea (31.6%), vomiting (8.7%), fatigue (6.1%), and anemia (5.6%), for D/T were: nausea (55.3%), anemia (15.2%), fatigue (6.9%), neutropenia (23.5%), thrombocytopenia (12.4%), febrile neutropenia (6%), stomatitis (6%), and PPE syndrome (6%). ORR for C was 4.3% including a CR, vs 10.9% ORR for D/T. Median survival (MS) was 8.5 mos for C, 13.6 mos for D/T (p=0.0001). Median progression-free survival was 2.3 mos for C and 4.4 mos for D/T, (p=0.0001). D/T MS was 14.2/10.8 mos, respectively. Conclusions: Canfosfamide did not meet the primary endpoint. C demonstrated single agent activity in P refractory or resistant OC and was well tolerated. C in combination with standard agents in less heavily treated OC trials are in progress.

No significant financial relationships to disclose.

Maintenance erlotinib (E) following first-line treatment with docetaxel, carboplatin and erlotinib in patients with ovarian carcinoma

5560

Background: We previously reported on the feasibility of a combination of erlotinib (E; a HER1/EGFR tyrosine-kinase inhibitor) with platinum and taxane chemotherapy (Agarwal et al, Ann Oncol 2004) and reported that almost 1/3rd patients (pts) stopped E during treatment due to toxicities which included intolerable rash, diarrhea and neutropenic sepsis. This update reports on those pts who then received E at 150mg daily as maintenance treatment. Methods: 48 pts initially received docetaxel (75 mg/m2) and carboplatin (AUC 5) following surgery q21d for 6 cycles in combination with daily oral E (50–100 mg) in dose-escalating cohorts. After 6 cycles, pts could continue with E alone (150mg/d) until progression. Results: 27 pts (56%) continued E beyond chemotherapy for a median of 8.6 months (mo) (range 2.3–32.5 mos), and 23/27 were escalated as planned to daily. Subsequently, 12/27 (44%) pts had their dose either reduced or interrupted for toxicity, which was cutaneous in 10/12 (83%) pts. However, the incidence of = grade 2 toxicity was low apart from alopecia (24%), rash (18%) and fatigue (15%). During follow-up (36 mo) 22/27 (81%) pts stopped E due to progressive disease and 3/27 (11%) stopped E due to cutaneous toxicity. Only 2/27 (7%) pts continue to receive E without evidence of progression, both at doses less than 150mg, again due to cutaneous toxicity. Pts receiving maintenance E had a median progression-free survival of 14.8 mo (95% CI 12.6–17.1 mo) and median overall survival 37.0 mo (95% ci 31.6–42.4 mo); for all pts these figures were 12.5 mo (95% CI 9.1–15.9 mo) respectively, and 37.0 mo (95% ci 27.3–46.7 mo). Conclusions: Maintenance E at 150mg daily following initial treatment with E plus docetaxel-carboplatin for ovarian carcinoma is associated with cutaneous toxicity which limits the dose and duration of treatment in a proportion of pts. The potential benefit of this approach can only be addressed in a randomized trial, and this is now underway under the auspices of the EORTC. A parallel translational study will examine the possibility that patients most likely to benefit can be predicted by molecular tumor analysis.

[Table: see text]

A phase I trial of the selective oral cyclin-dependent kinase inhibitor seliciclib (CYC202; R-Roscovitine), administered twice daily for 7 days every 21 days

Seliciclib (CYC202; R-roscovitine) is the first selective, orally bioavailable inhibitor of cyclin-dependent kinases 1, 2, 7 and 9 to enter clinical trial. Preclinical studies showed antitumour activity in a broad range of human tumour xenografts. A phase I trial was performed with a 7-day b.i.d. p.o. schedule. Twenty-one patients (median age 62 years, range: 39-73 years) were treated with doses of 100, 200 and 800 b.i.d. Dose-limiting toxicities were seen at 800 mg b.i.d.; grade 3 fatigue, grade 3 skin rash, grade 3 hyponatraemia and grade 4 hypokalaemia. Other toxicities included reversible raised creatinine (grade 2), reversible grade 3 abnormal liver function and grade 2 emesis. An 800 mg portion was investigated further in 12 patients, three of whom had MAG3 renograms. One patient with a rapid increase in creatinine on day 3 had a reversible fall in renal perfusion, with full recovery by day 14, and no changes suggestive of renal tubular damage. Further dose escalation was precluded by hypokalaemia. Seliciclib reached peak plasma concentrations between 1 and 4 h and elimination half-life was 2-5 h. Inhibition of retinoblastoma protein phosphorylation was not demonstrated in peripheral blood mononuclear cells. No objective tumour responses were noted, but disease stabilisation was recorded in eight patients; this lasted for a total of six courses (18 weeks) in a patient with ovarian cancer.

Participation of patients with gynecological cancer in phase I clinical trials: two years experience in a major cancer center

OBJECTIVES

This study aims at analyzing the clinico-demographic features that influence the recruitment of gynecological cancer (GC) patients to phase I trials. The possible clinical benefit to patients resulting from the participation in these trials has been also investigated.

METHODS

We performed a retrospective analysis of GC patients referred to the Phase I Unit of the Royal Marsden Hospital in Sutton (Surrey, UK), over 2 years.

RESULTS

Overall 68 GC patients were referred, and subsequently 32 (47.1%) enrolled. The percentage of patients enrolled increased as the distance to travel between the patient's residence and the hospital shortened (8.3% through 47.8% to 60.8%, for travel time >2, 1-2 or < or =1 h, respectively; p=0.008). Better performance status (PS) was found to be associated with higher enrollment rate with percentages increasing from 0 through 51.2 to 58.8 in cases with PS> or =2, PS=1, PS=0, respectively (p=0.015). Among the biochemical parameters, only hepatobiliary dysfunction was found to be associated with lower enrollment (p=0.012). Minimal response/disease stabilization was observed in 11 patients (34.4%). An increased median survival following the first visit was observed in patients enrolled compared to those not enrolled (8 versus 4 months, respectively, p=0.0055). In the multivariate analysis, only PS and enrollment in trials retained an independent prognostic role (p=0.031 and p=0.040, respectively).

CONCLUSIONS

This study, suggesting liver function and PS as important factors influencing the recruitment of GC patients to phase I trials could guide referral of patients to phase I Units. Moreover, the practical limitations imposed by long distance travel, together with the potential clinical benefit due to the participation to these trials, should encourage more investigators to develop phase I units in major cancer centers.