Search

Background The objective of this study was to report on a retrospective series of patients with epithelioid hemangioendothelioma (EHE) who received treatment with sirolimus within the Italian Rare Cancer Network. Methods From January 2005, 38 adult patients with advanced EHE received continuous-dosing sirolimus, 5 mg daily, until they developed either toxicity or disease progression. Disease progression in the 6 months before the start of treatment was required. Each pathologic diagnosis was reviewed. The daily dose of sirolimus was adjusted based on plasma levels. Response was retrospectively assessed by local investigators using Response Evaluation Criteria in Solid Tumors, version 1.1 (RECIST). Survival was estimated using the Kaplan-Meier method. Results All 38 patients (WW Domain Containing Transcription Regulator 1 [WWTR1]-positive, n = 37; transcription factor E3 [TFE3]-positive, n = 1) had disease progression before starting sirolimus (at baseline, 13 of 38 patients had the presence of serosal effusions and systemic symptoms). Thirty-seven patients were evaluable for response (there was 1 early interruption). The best RECIST responses were a partial response in 4 patients (10.8%), stable disease in 28 patients (75.7%), and disease progression in 5 patients (13.5%). At a 41.5-month median follow-up (interquartile range [IQR], 23.9-56.8 months), the median PFS was 13 months (95% CI, 3.7 months to not estimated [NE]), and the median OS was 18.8 months (95% CI, 10.6 months to NE). In patients who had serosal effusions at baseline, the median PFS was 4.8 months (IQR, 3.5-11.7 months), and the median OS was 10.6 months (IQR, 5.1-13.0 months), compared with 47.8 months (IQR, 11.4 months to NE) and 47.8 months (IQR, 15.7 months to NE), respectively, in patients without serosal effusions. Overall, sirolimus was fairly well tolerated, with 10 patients reporting irregular menstruation/ovary disfunction. Conclusions The current results confirm that sirolimus is active in EHE, leading to prolonged stabilization in most patients who present without serosal effusions. Serosal effusions are confirmed as an unfavorable prognostic sign associated with short survival, and sirolimus displays limited activity in this subgroup.

Background This study aimed to review the activity of cytotoxic chemotherapy in patients with inflammatory myofibroblastic tumors (IMTs) treated at nine European sarcoma reference centers. Materials and Methods Patients of any age, with histologically proven IMT, treated with anthracycline-based methotrexate plus/minus vinorelbine/vinblastine (MTX-V) or other chemotherapeutic regimens between 1996 and 2018 were retrospectively reviewed. Diagnosis was confirmed at the local level by an expert pathologist. Response was retrospectively assessed by local investigators by RECIST v1.1. Progression-free survival (PFS), relapse-free survival (RFS), and overall survival (OS) were computed by Kaplan-Meier method. Results Thirty-eight patients were included. Twenty-five patients (8 localized, 17 advanced disease) received an anthracycline-based regimen; 21 were evaluable for response. Overall response rate (ORR) was 10/21 (47.6%). At a 70.8-month median follow-up (FU), median RFS and median OS were not reached (NR) in patients with localized disease; median PFS and median OS were 6.3 (interquartile range [IQR]: 1.9–13.4) and 21.2 (IQR: 7.7–40.7) months in patients with advanced disease. Thirteen patients received MTX-V (4 localized, 9 advanced disease), all evaluable for response. ORR was 7/13 (53.8%). At a 56.6-month median FU, median RFS and median OS were 42.5 (IQR: 12.9–61.2) months and NR (no death events) in patients with localized disease, and NR (IQR: 24.9 to NR) and 83.4 months (IQR: 83.4 to NR) in patients with advanced disease. In the “other-regimens group,” responses were seen in 3/4 patients treated with oral cyclophosphamide and 1/2 with docetaxel/gemcitabine. Conclusion Anthracycline-based and MTX-V regimens are very effective in IMT, with a similar ORR in both groups. MTX-V achieved a prolonged disease control. Responses were also seen with oral cyclophosphamide and docetaxel/gemcitabine, but few patients were treated with these schedules. Implications for Practice Inflammatory myofibroblastic tumor (IMT) is an ultrarare sarcoma with known sensitivity to anaplastic lymphoma kinase (ALK) inhibitors in ALK-fused cases, although ALK inhibitors are not licensed in the disease. The current knowledge on the activity of cytotoxic chemotherapy is limited. This multi-institutional retrospective study on pediatric and adult patients with IMT shows that cytotoxic chemotherapy, and in particular anthracycline-based and methotrexate plus/minus vinorelbine/vinblastine regimens, represents a treatment option and can be considered in IMT patients irrespectively from ALK status. This study provides a benchmark for future studies on new medical therapies.

The treatment of patients with recurrent and/or metastatic (R/M) salivary gland adenoid cystic carcinoma (ACC) remains an unmet need.Patients with R/M disease with a history of clinical or symptomatic disease progression within 6 months and a maximum of 1 previous line of chemotherapy or a multiple kinase inhibitor received oral lenvatinib at a dose of 24 mg/day. The primary endpoint was the objective response rate; secondary endpoints included quality of life (QOL) (according to the European Organization for Research and Treatment of Cancer Quality of Life Questionnaire-Core 30 Items [EORTC QLQ-C30] and the European Organization for Research and Treatment of Cancer Quality of Life Questionnaire-Core Module Head and Neck Module [EORTC QLQ-HN35]), progression-free survival and overall survival, duration of response, and toxicities.Twenty-eight patients with R/M ACC were enrolled. Among 26 evaluable patients, 3 partial responses (11.5%) were reported. Target lesion reductions between 23% to 28% were observed in 4 of 20 patients with stable disease. Treatment-related adverse events were frequent (all grades, 96%; grade≥3 in 50% of cases according to version 4.03 of the National Cancer Institute Common Terminology Criteria for Adverse Events). The dose of lenvatinib was reduced in 24 patients, whereas in 21 patients the dose was reduced within the first 12 weeks and 4 patients maintained the full dose throughout treatment. The QOL deteriorated between baseline and 6 months with regard to Fatigue and Dry Mouth. There was no evidence of changes in Swallowing and Physical Functioning. At a median follow-up of 29 months, 2 patients remained on treatment, 10 patients were off protocol for disease progression and were alive with disease, and 14 patients had died of disease progression. The median overall survival, progression-free survival, and duration of response were 27 months, 9.1 months, and 3.1 months, respectively.Lenvatinib appears to have modest activity in ACC. Toxicities are common but manageable and QOL was found to deteriorate in some domains.

[Background] This observational, retrospective effort across Europe, US, Australia, and Asia aimed to assess the activity of systemic therapies in EHE, an ultra-rare sarcoma, marked by WWTR1-CAMTA1 or YAP1-TFE3 fusions., [Methods] Twenty sarcoma reference centres contributed data. Patients with advanced EHE diagnosed from 2000 onwards and treated with systemic therapies, were selected. Local pathologic review and molecular confirmation were required. Radiological response was retrospectively assessed by local investigators according to RECIST. Progression free survival (PFS) and overall survival (OS) were estimated by Kaplan-Meier method., [Results] Overall, 73 patients were included; 21 had more than one treatment. Thirty-three patients received anthracyclines regimens, achieving 1 (3%) partial response (PR), 25 (76%) stable disease (SD), 7 (21%) progressive disease (PD). The median (m-) PFS and m-OS were 5.5 and 14.3 months respectively. Eleven patients received paclitaxel, achieving 1 (9%) PR, 6 (55%) SD, 4 (36%) PD. The m-PFS and m-OS were 2.9 and 18.6 months, respectively. Twelve patients received pazopanib, achieving 3 (25%) SD, 9 (75%) PD. The m-PFS and m-OS were.2.9 and 8.5 months, respectively. Fifteen patients received INF-α 2b, achieving 1 (7%) PR, 11 (73%) SD, 3 (20%) PD. The m-PFS and m-OS were 8.9 months and 64.3, respectively. Among 27 patients treated with other regimens, 1 PR (ifosfamide) and 9 SD (5 gemcitabine +docetaxel, 2 oral cyclophosphamide, 2 others) were reported., [Conclusion] Systemic therapies available for advanced sarcomas have limited activity in EHE. The identification of new active compounds, especially for rapidly progressive cases, is acutely needed.

Purpose: The mTOR complex C1 (mTORC1) inhibitor everolimus in combination with the aromatase inhibitor exemestane is an effective treatment for patients with hormone receptor—positive (HR+), HER2-negative (HER2−), advanced breast cancer (HR+/HER2− aBC). However, everolimus can cause hyperglycemia and hyperinsulinemia, which could reactivate the PI3K/protein kinase B (AKT)/mTORC1 pathway and induce tumor resistance to everolimus. Experimental Design: We conducted a multicenter, retrospective, Italian study to investigate the impact of baseline and on-treatment (i.e., during first 3 months of therapy) blood glucose levels on progression-free survival (PFS) in patients with HR+/HER2− aBC treated with everolimus-exemestane. Results: We evaluated 809 patients with HR+/HER2− aBC treated with everolimus-exemestane as any line of therapy for advanced disease. When evaluated as dichotomous variables, baseline and on-treatment glycemia were not significantly associated with PFS. However, when blood glucose concentration was evaluated as a continuous variable, a multivariable model accounting for clinically relevant patient- and tumor-related variables revealed that both baseline and on-treatment glycemia are associated with PFS, and this association is largely attributable to their interaction. In particular, patients who are normoglycemic at baseline and experience on-treatment diabetes have lower PFS compared with patients who are already hyperglycemic at baseline and experience diabetes during everolimus-exemestane therapy (median PFS, 6.34 vs. 10.32 months; HR, 1.76; 95% confidence interval, 1.15–2.69; P = 0.008). Conclusions: The impact of on-treatment glycemia on the efficacy of everolimus-exemestane therapy in patients with HR+/HER2− aBC depends on baseline glycemia. This study lays the foundations for investigating novel therapeutic approaches to target the glucose/insulin axis in combination with PI3K/AKT/mTORC1 inhibitors in patients with HR+/HER2− aBC.

Contains fulltext : 220839.pdf (Publisher’s version ) (Closed access) BACKGROUND: Intimal sarcoma (InS) is an exceedingly rare neoplasm with an unfavorable prognosis, for which new potentially active treatments are under development. We report on the activity of anthracycline-based regimens, gemcitabine-based regimens, and pazopanib in patients with InS. METHODS: Seventeen sarcoma reference centers in Europe, the United States, and Japan contributed data to this retrospective analysis. Patients with MDM2-positive InS who were treated with anthracycline-based regimens, gemcitabine-based regimens, or pazopanib between October 2001 and January 2018 were selected. Local pathological review was performed to confirm diagnosis. Response was assessed by RECIST1.1. Recurrence-free survival (RFS), progression-free survival (PFS) and overall survival were computed by Kaplan-Meier method. RESULTS: Seventy-two patients were included (66 anthracycline-based regimens; 26 gemcitabine-based regimens; 12 pazopanib). In the anthracycline-based group, 24 (36%) patients were treated for localized disease, and 42 (64%) patients were treated for advanced disease. The real-world overall response rate (rwORR) was 38%. For patients with localized disease, the median RFS was 14.6 months. For patients with advanced disease, the median PFS was 7.7 months. No anthracycline-related cardiac toxicity was reported in patients with cardiac InS (n = 26). For gemcitabine and pazopanib, the rwORR was 8%, and the median PFS was 3.2 and 3.7 months, respectively. CONCLUSION: This retrospective series shows the activity of anthracycline-based regimens in InS. Of note, anthracyclines were used in patients with cardiac InS with no significant cardiac toxicity. The prognosis in patients with InS remains poor, and new active drugs and treatment strategies are needed.

Aim During the last years, the role of sentinel lymph node mapping (SLNM) for endometrial cancer (EC) surgical treatment has increased in popularity. However, several controversies remain about different technical steps of SLNM. Thus, a randomised control trial was designed to compare cervical (CI) and hysteroscopic (HI) indocyanine green (ICG) injection for SLNM of newly diagnosed EC undergoing surgical staging. The primary end-point of the study was to compare these two techniques in terms of para-aortic detection rate. Methods Patients with apparent stage I or II histologically confirmed EC undergoing surgery were included in the study. This randomised trial distinguished patients in two study groups according to two different techniques of ICG SLNM: CI versus HI injection. Patients who met the inclusion criteria were randomly assigned to CI or HI injection in a 1:1 ratio. The central randomisation system allocated patient randomisation numbers sequentially in the order in which the patients were enrolled. This randomised trial was not blinded for either patients or the surgeons. Results From March 2017 until April 2019, a total of 165 patients were randomised in this study: 85 (51.5%) in the CI group and 80 (48.5%) in the HI group. After randomisation, 14 (8.5%) patients were excluded from the study. Finally, 151 patients were included in the analysis: 82 (54.3%) in the CI group and 69 (45.7%) in the HI group. Hysteroscopy injection shows an ability to detect Sentinel nodes (SNLs) in the para-aortic area of about 10% greater compared with CI injection, although this difference did not reach statistical significance. The HI technique was superior in detecting isolated para-aortic SLNs (5.8% Versus 0%). The CI injection was correlated with higher SLN detection rates at the pelvic level compared with HI injection. Pelvic and overall detection was higher in the CI group. Conclusions The present study supports the adoption of CI instead of HI injection because the former allows better identification of sentinel nodes (especially in the pelvic area). Detection of SLN in the para-aortic area was slightly higher in patients receiving a HI injection, but the difference with the CI route was not statistically significant.

Background Distant metastases in adenoid cystic carcinoma (ACC) are common. There is no consensus on the management of metastatic disease because no therapeutic approach has demonstrated improvement in overall survival (OS) and because of prolonged life expectancy. The aim of this study is to build and validate a prognostic nomogram for metastatic ACC patients. Methods The study end-point was OS, measured from the date of first metastatic presentation to death/last follow-up. A retrospective analysis including metastatic ACC patients was performed to build the prognostic nomogram at the INT (Milan, Italy). The model was validated on an independent cohort of patients with similar characteristics treated at Leuven (Belgium). Outcome data and covariates were modelled by resorting to a random forest method. This machine-learning approach was used to guide and benchmark the subsequent use of more conventional modelling methods. Cox model performance was assessed in terms of discrimination (Harrell's c-index). Results Two hundred ninety-eight patients with metastatic ACC (testing set 259 INT, validation set 39 Leuven) were studied. Akaike Information Criterion–based backward selection yielded a 5-factor model showing a bias-corrected c-index of 0.730. Five independent prognostic factors were found: gender, disease-free interval and presence of lung, liver or bone metastases. Nomogram discrimination in the validation series was c = 0.701. Conclusion This retrospective analysis allowed the building of an externally validated prognostic nomogram. This tool might help clinicians to discriminate patients requiring prompt management from who can benefit from a ‘watchful waiting’. In addition, the nomogram might be useful to stratify patients in clinical trials.

11558 Background: R showed antitumor activity in a PDX model of dedifferentiated SFT (D-SFT), inducing a superior tumour growth inhibition than with other antiangiogenic agents (A), such as pazopanib (P) and axitinib (A). The efficacy of P and A in patients (pts) with advanced typical- (T-)/ malignant- (M-)SFT has been already confirmed in phase 2 clinical trials, but not in D-SFT. An exploratory phase 2 study was designed to investigate the activity of R in advanced SFT. Methods: An investigator-initiated exploratory phase 2 trial was started in December 2015 at the Istituto Nazionale Tumori, Milan, Italy, to evaluate the activity of R, 160 mg OD, 3 weeks on/1 week off, until progression or limiting toxicity. in > 18 years old pts with advanced SFT. The target sample size was 16 evaluable pts; at least 3 responses were requested to reject the null hypothesis of 5% in favour of an alternative hypothesis of 30% (with type-I and type-II error rates fixed at the 10% level). Eligible pts had to have evidence of progression. Prior treatment with A was allowed. Centralized pathologic review was performed, distinguishing T-SFT, M-SFT and D-SFT subtypes. The primary end-point was the overall response rate (ORR) by Choi. Secondary end-points were ORR by RECIST, progression-free survival (PFS), overall survival (OS). Results: Enrolment was completed in February 2021. Eighteen pts were enrolled (D-SFT = 4; M-SFT = 13; T-SFT = 1). Four pts were naïve, 14 were pre-treated [12 with antiangiogenics (4 with > 1 prior antiangiogenic line); 11 with cytotoxic agents]. Three pts are ongoing, 13 completed their treatment (11 = progression, 1 = toxicity, 1 = other). Fourteen pts are evaluable for response by Choi and RECIST (1 = screening failure; 1 = early discontinuation for toxicity before radiologic assessment; 2 = too early). A definitive dose reduction was required in 5 of 14 evaluable (35.7%) pts. The ORR by Choi was 42.9% (exact binomial 95% Confidence Interval [CI]: 17.7%-71.1%), with 6/14 (42.9%) partial responses (PR), 5/14 (35.7%) stable disease (SD) and 3/14 (21.4%) progressions (PD). Best responses by RECIST were: 1/14 (7.1%) PR, 10/14 (71.4%) SD, 3/14 (21.4%) PD. 5/6 pts responsive by Choi were pre-treated with another antiangiogenic. No responses were seen in the 3 D-SFT pts. At a m-FU of 23 months, m-PFS by Choi was 3.68 (IQR: 2.73-8.54) months, with 23.4% pts progression free at 1 year. m-PFS by Choi in responsive pts was 5.62 (IQR: 2.89 – 8.54) mos. Median OS was 15.7 (IQR: 7.35-not reached) months. Conclusions: R did not show a higher activity in D-SFT compared to P and A. The response rate was in the range observed with other A, but m-PFS was shorter. This may be due to discrepancies in pt populations and a high-rate of dose reductions with R. However, responses to R were seen also in pts previously treated with other A and almost one fourth of pts benefited from R for more than a year. Clinical trial information: 2015-002629-21.

11570 Background: In recent years evidence of long term stabilization and spontaneous regression of primary sporadic DT resulted in a paradigm shift towards more conservative approaches. We present herein the results of an observational study aimed at prospectively assess the behavior of primary sporadic DT initially managed by active surveillance and its relation to CTNNB1 mutational status. Methods: This is an Italian prospective, multicenter, observational study (NCT 02547831) to evaluate primary sporadic DT behavior in patients (pts) >16 years with primary naive or incompletely resected and measurable disease, at any site, with CTNNB1 mutational status available. Pts were assessed by contrast enhanced MRI or CT scan at baseline, at 3, 6, 9, 12 months, then every 6 months until the third year. Tumor changes were assessed by RECIST. In case of dimensional or symptomatic progression active treatment could be proposed on an individualized basis CTNNB1 mutational status was obtained in all patients by Sanger and deep sequencing was performed in wild-type cases. Primary end-point was progression-free survival (PFS) at 3 yrs. Treatment-free survival (TFS) was also analyzed. PFS and TFS were calculated using survival analysis methods, including Kaplan-Meier plots, log-rank test for survival curves comparison, and Cox proportional-hazard multivariable regression (age, size, anatomic site and CTNNB1 mutational status). Results: From April 2013 to February 2018 108 pts were included (82% female, 18% male); median age was 39 (interquartile range, IQR 34-48) and median size 50 mm (IQR, 40-80 mm). Tumor location was: 4/108 (4%) = head&neck, 25/108 (23%) = trunk, 59/108 (55%) = abdominal wall, 3/108 (3%) = intra-abdominal, 17/108 (16%) = extremities. CTNNB1 mutations were as follow: T41A 54/108 (50%), S45F 13/108 (12%), WT 20/108 (19%), other mutations 21/108 (19%). At a m-FU of 32.3 months, the 3-year PFS was 54.5% (95% CI, 44.9%-66.1%). 42/108 (39%) pts showed a RECIST progression, with equal distribution among the different anatomic sites. None of the variables analyzed were associated to PFS. Spontaneous regression was initially observed in 27/108 (25%) patients, while it followed dimensional progression in another 33/108 (30%). 35/108 (32%) pts received active treatment, 18/42 (43%) after RECIST progression and 17/66 (26%) after symptomatic progressions. TFS at 36 months was 65.9% (95% CI, 57.3%-75.9%). S45F mutation, larger tumor size and extremity location were associated to a shorter TFS. No dimensional or symptomatic progression were observed after 36 months of FU. Conclusions: Active surveillance was marked by spontaneous regressions in 60/108 (55%) pts. An active treatment was needed in 32%. If no events occur after 3 yrs of FU, more relaxed FU schedules can be considered given the low risk of subsequent progression. Attention should be paid to patients with DT located to the extremity and/or carrying a S45F mutation. Clinical trial information: NCT 02547831.

11065 Background: To report on a multi-institution retrospective study on the activity of anthracycline-based (Ab) and gemcitabine-based (Gb) regimens as well as pazopanib (P) in patients with advanced epithelioid sarcoma (ES) treated within 16 sarcoma reference centres in Europe, US and Japan. Methods: Patients with a histologically confirmed diagnosis of locally advanced/metastatic ES were selected. Classic and distal subtypes were defined based on morphology (WHO 2014). INI1 expression is under evaluation.Response was evaluated by RECIST.Progression-free survival (PFS) and overall survival (OS) were computed by Kaplan-Meier method. Results: Ninety ES patients were identified (Table 1). They were treated with Ab (72), Gb (30) and P (20); 25 pts received more than one treatment. The median follow-up for Ab, Gb and P groups was 32, 24 and 22 months, respectively. The response rate (RR) for Ab was 25% (95% CI 16%-37%), with a median PFS and OS of 6 and 17 months. The RR for Gb was 23% (95% CI 10% - 42%), with 1 complete response and a median PFS and OS of 5 and 20 months. In the P group, no objective responses were reported, and median PFS and OS were 3 and 9 months. A non-statistically significant trend towards a greater RR in proximal than classic subtype was seen in both Ab (27% vs 22%) and Gb (30% vs 13%) groups. Conclusions: This retrospective series, the largest currently available, confirms the activity of Ab and Gb in ES, with a similar RR and PFS in both groups. In this population, the value of P seems limited. These data may serve as a benchmark for trials of novel agents in ES. [Table: see text]

Purpose Obesity is a known risk factor for breast cancer and has been linked to increased risk of recurrence and death in breast cancer patients. Little is known about the predictive value of obesity. As endocrine therapy is widely used for breast cancer treatment worldwide, we aimed at correlating baseline body mass index (BMI) with clinical benefit derived from fulvestrant in postmenopausal women with advanced breast cancer. Methods We analyzed consecutive patients treated with fulvestrant in our center between January 2009 and March 2015. Patients were categorized as normal (BMI 18.5-24.9 kg/m2), overweight (BMI 25-29 kg/m2) and obese (BMI >30 kg/m2). The antitumor activity of fulvestrant was evaluated in terms of the clinical benefit rate (CBR). Results Seventy-five consecutive patients matched the eligibility criteria for analysis. Fulvestrant was administered as first-line therapy in 4 (5%) cases, as second line in 27 (36%) and as third line and beyond in 44 (59%) cases. According to BMI, 44 (59%) patients were classified as normal weight, 19 (25%) as overweight, and 12 (16%) as obese. No difference in estrogen receptor expression was found in relation to BMI. CBR was 53% overall, but rose to 70.5% in normal-weight patients and dropped to 31.6% and 25% in overweight and obese patients, respectively (pConclusions Increased BMI has a negative influence on treatment outcome. Even with the limitation of the relatively small sample size, it appears that patients of normal weight are 2.5-fold more likely to benefit from fulvestrant as overweight and obese patients.

Item does not contain fulltext Importance: Epithelioid sarcoma (ES) is an exceedingly rare malignant neoplasm with distinctive pathologic, molecular, and clinical features as well as the potential to respond to new targeted drugs. Little is known on the activity of anthracycline-based regimens, gemcitabine-based regimens, and pazopanib in this disease. Objective: To report on the activity of anthracycline-based regimens, gemcitabine-based regimens, and pazopanib in patients with advanced ES. Design, Setting, and Participants: Seventeen sarcoma reference centers in Europe, the United States, and Japan contributed data to this retrospective analysis of patients with locally advanced/metastatic ES diagnosed between 1990 and 2016. Local pathological review was performed in all cases to confirm diagnosis according to most recent criteria. Exposures: All patients included in the study received anthracycline-based regimens, gemcitabine-based regimens, or pazopanib. Main Outcome and Measures: Response was assessed by RECIST. Progression-free survival (PFS) and overall survival (OS) were computed by Kaplan-Meier method. Classic and proximal subtypes were defined based on morphology (according to 2013 World Health Organization guidelines). Results: Overall, 115 patients were included, 80 (70%) were men and 35 (30%) were women, with a median age of 32 years (range, 15-77 years). Of the 115 patients with ES, 85 were treated with anthracycline-based regimens, 41 with gemcitabine-based regimens, and 18 with pazopanib. Twenty-four received more than 1 treatment. Median follow-up was 34 months. Response rate for anthracycline-based regimens was 22%, with a median PFS of 6 months. One complete response (CR) was reported. A trend toward a higher response rate was noticed in morphological proximal type (26%) vs classic type (19%) and in proximal vs distal primary site (26% vs 18%). The response rate for gemcitabine-based regimens was 27%, with 2 CR and a median PFS of 4 months. In this group, a trend toward a higher response rate was reported in classic vs proximal morphological type (30% vs 22%) and in distal vs proximal primary site (40% vs 14%). In the pazopanib group, no objective responses were seen, and median PFS was 3 months. Conclusions and Relevance: This is the largest retrospective series of systemic therapy in ES. We confirm a moderate activity of anthracycline-based and gemcitabine-based regimens in ES, with a similar response rate and PFS in both groups. The value of pazopanib was low. These data may serve as a benchmark for trials of novel agents in ES.

Background We present the results of an academic phase 2 study on imatinib plus everolimus in patients who have progressive advanced chordoma. Methods In January 2011, 43 adult chordoma patients were enrolled in the study and received imatinib 400 mg/day and everolimus 2.5 mg/day until progression or limiting toxicity. Eligible patients had progressed in the 6 months before study entry. PDGFRB, S6, and 4EBP1 expression and phosphorylation were evaluated by way of immunohistochemistry and/or western blotting. The primary endpoint was the overall response rate (ORR) according to Choi criteria. Secondary endpoints were RECIST 1.1 response, progression-free survival (PFS), overall survival (OS), correlation between S6/4EBP1 phosphorylation and response. Results Thirteen of 43 patients were pretreated with imatinib. Among 40 of the 43 patients who were evaluable by Choi criteria, the best responses were 9 with partial response (ORR, 20.9%), 24 with stable disease (SD) (ORR, 55.8%), and 7 with progressive disease (ORR, 16.3%). Forty-two patients were evaluable by RECIST criteria, with 1 partial response (ORR, 2.3%), 37 stable disease (ORR, 86%), and 4 progressive disease (ORR, 9.3%). The median PFS according to Choi criteria was 11.5 months (range, 4.6-17.6 months), and 58.8% and 48.1% of patients were progression-free at 9 and 12 months, respectively. The median PFS by RECIST criteria was 14 months; the median OS was 47.1 months. When assessable, S6/4EBP1 was phosphorylated in a high and moderate/low proportion of tumor cells in responsive and nonresponsive patients, respectively. Toxicity caused a temporary and definitive treatment discontinuation in 60.5% and 30.2% of patients, respectively. Conclusions Imatinib plus everolimus showed a limited activity in progressing advanced chordoma. Interestingly, the amount of tumor cells activated for mammalian target of rapamycin effectors correlated with the response. Toxicity was not negligible.

Background Limited information is available regarding the use of androgen receptor (AR) immunohistochemical expression in muscle-invasive or metastatic urothelial carcinoma. We aimed to evaluate the frequency of AR expression by tumor cells (TC), its prognostic role, and its relationship with programmed cell-death ligand 1 (PD-L1) expression in these patients. Patients and Methods From September 2015 to January 2017, we collected tissue from patients who received platinum-based chemotherapy at our center. Immunohistochemistry for AR was performed (1% cutoff of TC). PD-L1 coexpression, by TC or immune cells (1% cutoff), was also analyzed. Molecular analysis of AR gene was performed by sequencing of exons 5 to 8 and by fluorescence in-situ hybridization analysis. Cox models for overall survival (OS), adjusted for stage, visceral metastases, and platinum type, were fitted. Results A total of 110 patients had tumor samples stained. Overall, 48 (43.6%) had AR-expressing TC: 19 (17.3%) had 1%-5% expression, 15 (13.6%) 5%-25% expression, and 14 (12.7%) > 25% expression. Among the latter, 7 had molecularly evaluated tumor tissue: no AR gene mutations or amplifications were found, but polysomy of Xq chromosome was seen. PD-L1 expression by TC and immunohistochemistry concordantly decreased with increasing levels of AR expression by TC. In Cox analyses, AR expression was not associated with OS, both on univariable ( P = .477) and multivariable ( P = .505) analyses. Conclusion AR is frequently expressed in patients with muscle-invasive and advanced urothelial carcinoma, and it does not seem to be prognostic for OS. The AR pathway is worthy of clinical studies to assess its synergistic action with anti–PD-L1 therapy.

512 Background: Advances in prognostic allocation of pts with PSCC receiving regional LND are needed, along with an assessment of the role of perioperative treatments. Methods: An international, multicenter, retrospective database was queried for all pts with PSCC who had received inguinal ± pelvic LND from 1980 to 2017. We classified or re-classified pts according to the 2009 TNM classification system. The primary endpoint was OS. Survival analysis methods were adopted for the purpose of detecting and modeling putative prognostic factors, and develop a prognostic stratification tool. The analysis was performed in two steps: first, we modeled outcome data and covariates by resorting to a random forest (RF) method (an “ensemble” machine-learning approach, with an ad hoc approach incorporating missing data imputation). Second, data was modeled using Cox proportional hazard regression. In addition, a Cox model was fit including pre-specified variables that were deemed to be clinically relevant. Results: There were 743 pts who received LND at seven referral centers from Europe, the USA, Brazil, and China. Of these pts, 689 were analyzed: 86 (12.5%) received neoadjuvant chemotherapy (NAC); 171 (24.8%) received adjuvant chemotherapy (AC), and 74 (10.7%) received adjuvant radiotherapy. The variables significantly associated with OS were age (p < 0.001), the pathologically involved/total removed LN ratio (p < 0.001), pN stage (overall p < 0.001) and NAC (HR: 1.58, 95%CI: 1.10-2.26, p = 0.013). NAC and AC were ineffective or detrimental in N1-2 patients, whereas they provided OS improvements in N3 patients. OS improvement reached statistical significance for AC in pelvic LN-positive pts (p = 0.046). Finally, we developed a nomogram predicting 12- and 24-month OS based on pre-specified variables (c-index: 0.75). The study is limited by its retrospective nature. Conclusions: We propose a post-treatment prognostic tool that can be offered as an aid to physicians to enhance decision-making, clinical research, and patient counseling after LND in PSCC. Administration of NAC and AC should be restricted to N3 patients.

426 Background: 6 cycles of platinum-based chemotherapy (CT) are conventionally targeted to treat locally advanced unresectable or metastatic urothelial carcinoma (UC). However, cisplatin is associated with significant cumulative toxicities, which render it challenging to deliver 6 cycles. Since this issue has not been investigated prospectively, we conducted a retrospective analysis. Methods: The Retrospective International Study of Invasive/Advanced Cancer of the Urothelium (RISC) database was used to conduct a retrospective analysis. The association of the number of cycles of platinum-based first-line CT with overall survival (OS) was investigated by a Cox regression utilizing multivariate analysis after controlling for previously recognized prognostic factors used in a nomogram ( Eur Urol, 2017). The primary analysis was a comparison of < 6 cycles vs. ≥6 cycles. Six-month landmark analysis was applied throughout, accounting for OS events. Additionally, we excluded patients (pts) receiving < 3 or > 9 cycles to reduce confounding due to early removal for toxicities, progression and patient decision and increased number of cycles due to response and pt-related factors. Results: Of 1020 pts available from RISC, 472 (cisplatin = 338, carboplatin = 134) were evaluable for the landmark analysis with 281 events. A total of 157 pts received 3-5 cycles (median 4) and 315 received 6-9 cycles (median 6). There was no significant difference between 3-5 vs. 6-9 cycles of platinum-based chemotherapy (HR 1.02, 95%CI: 0.77-1.33, p = 0.91). No significant interactions were observed with type of platinum (p = 0.09) and “completed planned CT” (p = 0.56). Comparison of 4 vs. 6 cycles (p = 0.57) and < 6 vs 6 vs 7-9 (p = 0.9) also yielded no significant differences for association with OS. No differential association was observed with survival for 3-5 vs. 6-9 cycles when examining by nomogram-defined risk group tertiles. Limitations of a hypothesis-generating retrospective analysis apply. Conclusions: 4 cycles of platinum based first-line CT appear adequate to treat advanced UC. The omission of excessive cycles will avoid unnecessary toxicities and facilitate better transition to second-line and switch maintenance therapy.

The identification of lung tumor-initiating cells and associated markers may be useful for optimization of therapeutic approaches and for predictive and prognostic information in lung cancer patients. CD133, a surface glycoprotein linked to organ-specific stem cells, was described as a marker of cancer-initiating cells in different tumor types. Here, we report that a CD133 + , epithelial-specific antigen-positive (CD133 + ESA + ) population is increased in primary nonsmall cell lung cancer (NSCLC) compared with normal lung tissue and has higher tumorigenic potential in SCID mice and expression of genes involved in stemness, adhesion, motility, and drug efflux than the CD133 − counterpart. Cisplatin treatment of lung cancer cells in vitro resulted in enrichment of CD133 + fraction both after acute cytotoxic exposure and in cells with stable cisplatin-resistant phenotype. Subpopulations of CD133 + ABCG2 + and CD133 + CXCR4 + cells were spared by in vivo cisplatin treatment of lung tumor xenografts established from primary tumors. A tendency toward shorter progression-free survival was observed in CD133 + NSCLC patients treated with platinum-containing regimens. Our results indicate that chemoresistant populations with highly tumorigenic and stem-like features are present in lung tumors. The molecular features of these cells may provide the rationale for more specific therapeutic targeting and the definition of predictive factors in clinical management of this lethal disease.

Purpose To explore whether the adoption of a systematic attempt to perform wider resections may lead to prognostic improvements in retroperitoneal soft tissue sarcoma (RSTS). Patients and Methods Two hundred eighty-eight consecutive patients who were surgically treated at a single referral center were analyzed. Because a shift toward a systematic, more aggressive surgical approach (ie, liberal en bloc resection of adjacent organs) was in place from 2002 onward, patients were divided in two groups accordingly. Overall survival, crude cumulative incidence (CCI) of local recurrence, and distant metastases were estimated. Univariable and multivariable analyses were carried out. Results Patients who underwent operation in the early period had a 5-year local recurrence rate of 48% compared with 28% for patients who were treated in the recent period. The number of distant metastases was greater in the recent group (22% v 13%), and overall survival was similar. In addition to the period of treatment, important independent determinants for local recurrence-free survival were histologic grade, histologic subtype, and radiation therapy. Overall, liposarcomas and grades 1 to 2 tumors had the greatest local benefit at 5 years. Conclusion In a single institution, the adoption of a policy of more liberal visceral en bloc resections was paralleled by greater local control. This benefit might translate into a prognostic improvement only on a longer follow-up for patients with a more indolent disease, whereas systemic failures seem to be the main problem in high-grade tumors. Radiation therapy could add some additional benefit to local outcome and possibly to survival.

Aims and background Hope is crucial for patients with cancer. We explored the determinants of hope in patients with cancer using a questionnaire administered over the course of 1 day to an unselected sample of patients at an Italian cancer center. Methods A team of oncologists, statisticians, and chaplains developed a questionnaire with medical, psychological, spiritual, and religious content. A cross-sectional study was conducted on 320 patients who answered the questionnaire. Results In the group of participants, 92.8% had a religious belief. Women, patients with limited formal education, and believers were more hopeful. Patients placed trust in God, their partners and children, scientific research, and doctors. On univariate and multivariate analysis, hope was found sensitive to patients' sharing their experiences with others (including family and friends), their positive perception of the people around them, and their relationship with doctors and nurses. Conclusions If validated in further studies, these results support the notion that a patient with cancer's sense of hope is sensitive to the quality of relationships with caregivers. This may be important to health care organization and resource allocation.

Purpose Dermatofibrosarcoma protuberans (DFSP) is a rare, low-grade, cutaneous sarcoma with autocrine overproduction of the platelet-derived growth factor (PDGF) β-chain from gene rearrangement as a key pathogenetic factor, now susceptible of molecular-targeted therapy. The aim of this retrospective analysis was to explore the outcome of patients with primary or recurrent DFSP. Patients and Methods Two hundred eighteen patients surgically treated at the Istituto Nazionale per lo studio e la cura dei Tumori (Milan, Italy) over 20 years were reviewed. Local relapse, distant metastasis, and survival were studied. Results One hundred thirty-six patients (62.4%) presented with a primary DFSP, while 82 patients (37.6%) had a recurrent disease. In the primary group, margins were microscopically positive in 11.8%, while in the recurrent group they were positive in 14.6% (P =.613). In the primary group, patients undergoing re-excision after inadequate previous surgery had residual disease in 62% of cases. Reconstructive surgery was needed in 30%, significantly more frequently in patients with a recurrence or a head and neck tumor. The crude cumulative incidence of local relapses was 4% at 10 years, and 2% at 10 years for distant metastases. No significant difference was found between primary and recurrent patients, as well as between positive and negative margins. Conclusion This being one of the largest mono-institutional series of DFSP, we confirm that long-term outcome is excellent, in terms of both local and distant control, after a wide excision with negative margins. Reconstructive surgery is often needed. Novel medical therapies will be of use in a limited subgroup of patients.

Background: The prognosis of patients (pts) with PSCC is primarily depending on the involvement of regional lymph nodes. Owing to the limited efficacy of chemotherapy, new drugs are warranted. Dacomitinib (Daco) is a potent, irreversible TKI of human EGFR/HER1, HER2 and HER4. Methods: In a phase 2 study (NCT01728233), pts received Daco 45 mg/d, orally, continuously. Inclusion criteria were SCC histology, and clinical stage N2-3 or M1; no prior chemotherapy was allowed. Computed tomography (CT) and PET/CT scans were repeated q8 weeks. Responding (R) pts with locally-advanced (LA) PSCC were offered radical surgery. The primary endpoint was the objective response-rate (ORR=CR+PR according to RECIST v1.1). Stopping rules based on the Bayesian posterior probability (PP) were set; the target was to demonstrate that the ORR exceeded 20%. Translational analyses on pre-Daco tumor samples included: EGFR and HER2 amplification, in-situ hybridization for HPV, and next generation sequencing. Results: From 06/13 to 010/16, 28 patients were treated. 8 (28.6%) had visceral metastases. 50% had pelvic and 60.7% clinically-involved bilateral LN. 1 CR + 8 PR were obtained (ORR=32.1%, 80% credibility interval 21.0-43.0%). The median follow up was 19.8 months (IQR: 6.3-25.7); 12-m PFS was 26.2% (95%CI: 13.2-51.9); 12-m OS was 54.9% (95%CI: 36.4-82.8). The median OS of LA pts was 20 months (IQR: 11.1-NR). The current Bayesian PP of exceeding the 20% ORR target is 92.3%. Grade 3 adverse events (skin rash) were seen in 3 pts (10.7%). The median drug exposure was 2.2 months. Tissue samples from 25 pts were analyzed. Only 2 pts were HPV+ (1SD, 1PD). EGFR ampl was found in 4 pts (1 CR, 1 PR, 2 SD) and it was confirmed in all post-Daco samples. TERT mutations (60%) were found in responders only, PI3K/mTOR pathway gene mutations in 42.9% R vs 8.3% non responders, IGF2R in 40% and 20%, respectively. Mutations were largely consistent between matched pre- and post-therapy samples in 7 evaluated pts. Conclusions: Daco was active in advanced PSCC and provided the best ORR with a single agent in PSCC. The biology underlying response to Daco seems to be independent from HER-pathway alterations. The frontline therapy can provide a suitable window of opportunity to test new drugs in PSCC. Note: This abstract was not presented at the meeting. Citation Format: Andrea Necchi, Salvatore Lo Vullo, Daniele Raggi, Patrizia Giannatempo, Nicola Nicolai, Giuseppina Calareso, Elena Togliardi, Maurizio Colecchia, Federica Perrone, Annunziata Gloghini, Luigi Mariani, Roberto Salvioni. First-line therapy with dacomitinib, an orally available pan-her tyrosine kinase inhibitor, for locally-advanced or metastatic penile squamous cell carcinoma: Results of an open label, single-arm, single-center, phase 2 study [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr CT139. doi:10.1158/1538-7445.AM2017-CT139

e16019 Background: Advanced UC treatment is being revolutionized by immunotherapy (IT). However, < 30% of patients (pts) will benefit from IT, and additional therapeutic targets are needed. Findings from analyses supporting an investigator-led, AR-based clinical research program are presented. Methods: From 09/2015, we collected the archival tissue blocks from pts who have been treated or consulted at our center and received platinum-based chemotherapy (CT) for MIUC or mUC. Immunohistochemistry (IHC) was performed on 3µm slides using mouse anti-human AR monoclonal Ab (Dako, clone AR441). ≥1% cutoff for AR+ IHC of tumor cells (TC) was used. AR-expressing pts were further stratified as follows: 1-5%, 5-25%, and 25-100%. Univariable (UVA) and multivariable (MVA) Cox models for overall survival (OS) were fitted, the latter adjusted for clinical stage (MIUC vs mUC), presence of visceral metastases, platinum CT type. OS was calculated from the date of first administration of CT. Exploratory PD-L1 co-expression, by TC or stroma (SP142 Ab, 1% cutoff), was analyzed in all pts. Results: 105 pts (46 MIUC and 59 mUC) had their tumor stained. 80 (76.2%) had bladder primary tumor, 85 (80.9%) had pure UC. 59 received cisplatin and 46 carboplatin. Overall, 37 pts (35.2%) had AR-expressing TC, 17 (16.2%) with ≥25% expression. Cox models were built on 93 pts with follow-up duration of ≥6 months. AR-expressing was equally distributed between MIUC and mUC pts (44.4% and 55.6%). On UVA, AR expression was not associated with OS, neither dichotomizing AR+/AR- pts (p = 0.477) nor after accounting for AR-positivity strata (p = 0.845). The same results were confirmed on MVA (p = 0.505 and p = 0.875). In AR+ vs. AR- pts, PD-L1 was equally expressed in the stroma (67.5 vs 57.9%) but less expressed by TC (39.4% vs 52%). Conclusions: AR is frequently expressed on TC of pts with MIUC and mUC, and AR expression does not seem to be associated with OS in these pts. AR pathway is worthy of clinical studies to synergistically act with anti-PD-L1 therapy. AR sequencing and FISH analysis in positive pts is ongoing, and a clinical program of a named-basis administration of antiandrogen therapy has started at our center.

414 Background: Therapeutic options for patients (pts) with chemoresistant GCT are limited and prognosis is poor. PZP is a potent and selective tyrosine kinase inhibitor (TKI) with distinct antiangiogenic activity. Methods: In Pazotest trial (NCT01743482), pts were given PZP 800 mg/day until disease progression (PD) or unacceptable toxicity. Eligibility included failure of ≥ 2 platinum-based CT, including high-dose (HD)-CT. Baseline serum tumor markers (STM), computed tomography and FDG-PET were repeated after 4 wks and q8 weeks thereafter. The primary endpoint was PFS (H0: 3-m PFS ≤ 10%, H1: ≥ 25%, α = 5%/β = 20%). PD was defined as rising STM, increasing size of nonteratomatous masses (RECIST v1.1), or the appearance of new lesions. Translational analyses included circulating IL8 levels and next generation sequencing of pre-PZP tissue from extreme responders. Results: 43 pts were enrolled from 05/2013 to 07/2016. 35 pts (81.4%) had nonseminoma, 6 (14%) primary mediastinal GCT. The number of failed CT regimen was: 2 (11.6%), 3 (51.2%), > 3 (37.2%). 53.5% had failed HDCT. 39 pts had elevated STM: 56.4% AFP, 43.6% HCG. Grade 3-4 adverse events were seen in 4 pts (increased AST/ALT). 26/39 pts (66.7%) had STM reduction (38.5% > 50% from baseline) after 4 wks. RECIST results: 1 partial response, 20 stable disease, 16 PD (6 not evaluable). 12 (27.9%) had a decreased FDG uptake. Median follow-up was 29.6 months. 3-m PFS was 12.8% (95%CI: 5.7-28.9), median PFS was 2.5 m (IQR: 1.0-3.0). 24-m OS was 14.2% (95%CI: 6.0-33.7). In pts with > 50% STM decline, 24-m OS was 24.1% (95%CI: 8.3-69.6). Univariably, STM was associated with OS (HCG vs AFP: HR: 2.73, 95%CI: 1.24-6.02, p = 0.042). Baseline and +4wks IL8 levels did not predict for PFS/OS. In 2 responding pts, new mutations were found ( FLT3, G846V and RAD50 E448Q classified as damaging; TSC2, R197M and TNF A16V classified as benign). Conclusions: PZP was well tolerated and showed predictable activity, with early but short-living STM decline. Long term survival was obtained in a significant proportion of pts. Based on the present results, PZP may be investigated in earlier disease stages as an optimal bridging therapy preceding salvage curative treatment. Clinical trial information: NCT01743482.

345 Background: Despite cisplatin-based chemotherapy (CT) is recommended in MIUBC, an unmet need is to identify new drugs or combinations to improve the outcomes. SGC combination was evaluated in an open-label, single-arm, phase 2 trial (NCT01222676). Methods: After TURB, pts with T2-T4a N0 MIUBC received 4 cycles of cisplatin 70 mg/m2 d1, gemcitabine 1000 mg/m2d1 and 8, q3 wks. Sorafenib 400 mg q12h was administered daily from day 1 until radical cystectomy. In a Simon’s 2-stage design, the primary endpoint was pathologic complete responses (pT0). Residual carcinoma in situ was considered pT0. ITT analysis was applied. Statistical hypothesis assumed H0: ≤0.20 and H1: ≥0.40 (α and β of 5% and 10%). ERCC1 immunohistochemistry (IHC) and next-generation sequencing (NGS) of TURB tissue, and measurement of baseline circulating VEGF levels, were planned. Results: From 04/2011-06/2016, 45 pts were enrolled. 28 pts (62.2%) had macroscopical residual disease after TURB, 17 (37.8%) presented with clinical stage T3-4. pT0 was obtained in 19 pts (42.2%, 95%CI: 27.6-57.9); pT≤1 in 24 (53.3%, 95%CI: 37.9-68.3). After median follow-up of 35 months, median PFS was not reached (IQR: 29.4-NE), like median OS (IQR: 30.3-NE). Pathological response (pT0 vs pT≤1 vs other) predicted for both PFS (p=0.015) and OS (p=0.046). Hematologic (hem) G3-4 adverse events (AE): platelet 28.9%, neutrophils 15.6%, Hb (4.4%). Extra-hem AEs were seen in 9 pts (20%). 24 (53.3%) pts needed S temporary interruption (5 cases discontinuation). Tumor samples were available from 23 pts for ERCC1-IHC, 24 for NGS: ERCC1 IHC expression was associated with non response (pT≥2, p=0.033). ERBB2/ERBB3 mutations were found only in responders (pT≤1, 20%). PIK3CA/AKTmutations (missense, and newly-identified mutations) were more frequent in pT≥2 pts (35.7% vs 20%). Baseline VEGF levels did not predict for PFS/OS. NGS of the remaining pts of the trial is ongoing. Conclusions: SGC combination was active and effective in MIUBC. Translational analyses are providing information to develop combination of CT with new more potent and selective TKIs, possibly in more selected pts. Clinical trial information: NCT01222676.

399 Background: The prognosis of patients (pts) with PSCC is primarily depending on the involvement of regional lymph nodes (LN). Owing to the limited efficacy of chemotherapy (ChT), new drugs are warranted. Daco is a potent, irreversible TKI of human EGFR/HER1, HER2 and HER4. Methods: In a phase 2 study (NCT01728233), pts received Daco 45 mg/d orally continuously. Inclusion criteria were SCC histology, and clinical stage N2-3 or M1; no prior ChT was allowed. Computed tomography (CT) and PET/CT scans were repeated q8 wks. Responding pts with LA-PSCC were offered radical surgery. The primary endpoint was the objective response-rate (ORR = CR+PR according to RECIST v1.1). The study was conducted according to a Bayesian design, based on predictive probability; the target was to demonstrate that the ORR exceeded 20%. Translational analyses on pre-Daco tumor samples included: EGFR and HER2 amplification (ampl), in-situ hybridization for HPV, and next generation sequencing. Results: From 06/13 to 09/16, 26 pts were treated. 7 (26.9%) had visceral metastases. 50% had pelvic and 57.7% clinically-involved bilateral LN. 1 CR + 7 PR were obtained (ORR = 30.4%, 95% credibility interval 14.9-48.6%). Median follow up = 16.6 months; 12-m PFS = 24.1% (95%CI: 11.1-52.3); 12-m OS = 50.7% (95%CI: 31.7-80.9). The current Bayesian posterior probability of exceeding the 20% ORR target is 89%. Tissue samples from 23 pts were analyzed. Only 2 pts were HPV+ (1SD, 1PD). EGFR ampl was found in 4 pts (1 CR, 1 PR, 2 SD) and it was confirmed in all post-Daco samples. TERT mutations (60%) were found in responders (R) only. Mutations in genes potentially associated with resistance to EGFR inhibitors ( HRAS, BRAF, PIK3CA, PTEN, STK11) were found in 47% non responders (NR) vs 25% R. HRAS and BRAF mutations segregated with NR (20%). Additional unfrequent mutations of druggable targets were found: RAC1, TSC1 and mTOR in R, FGFR2, RAD50, PDGFRB and FLT4 in NR, IGF2R in both. Conclusions: Dacomitinib was active in HPV-negative, advanced PSCC. Among multiple altered pathways, we were able to delineate the possible molecular profile of responders. An expansion cohort to confirm the translational findings is planned. Clinical trial information: NCT01728233.

e16026Background: The available prognostic models (PM) for OS of pts with mUC were derived from clinical trial populations of cisplatin (cis)-treated pts. We sought to validate and improve these PM...

382 Background: Progress in developing effective salvage therapies for UC is warranted. AAK overexpression has been described in UC and spindle checkpoint (SC) dysregulation is a common feature. Alisertib is an orally available, selective TKI of AAK. Methods: A single-group, Ph2 trial was conducted with Alisertib 50 mg orally BID for 7 days, with 14d rest until disease progression (PD) (NCT01653028). The primary endpoint was RECIST 1.1 objective response-rate (ORR), and ≥ 3/20 responses were required (H0 ≤ 5%, H1 ≥ 20%, α = 5%/β = 20%) to move to a randomized trial of paclitaxel plus either Alisertib or placebo. Eligibility included failure of ≤ 2 platinum-based regimens. Cox regression analyses were done. Targeted sequencing of pts is being performed through the Ion AmpliSeq Comprehensive Cancer Panel (Life Technologies). Results: From 10/2014 to 04/2015, 22 pts were enrolled (20 evaluable for response), 8 (36.4%) in 2nd line and 14 (63.6%) > 2ndline. 8 (36.4%) had an ECOG-PS 1-2, 3 (13.6%) had Hb < 10mg/dl, 9 (40.9%) had liver metastases. 2 PR (ORR: 10%), 7 stable disease (SD) and 11 PD were recorded. Median FUP was 6.3 mos (IQR: 4.5-8.0). 6m-PFS was 18.2% (95%CI: 7.5-44.1), including 2 pts with 9.7 and 9.5 mos FUP (still ongoing, both > 2nd line). Median PFS was 1.8 mos (IQR: 1.3-3.4), but median OS was not reached (6m-OS: 63.6%, 95%CI: 46.4-87.3). Hb < 10gr/dl was significantly associated with shorter PFS and OS multivariably (p = 0.031 and p = 0.033). AAK-IHC expression did not correlate with outcome. Tissue of pt with 9.7 month SD harbored a missense mutation of mTOR (E1813D*) and the nonsense mutation Q527STOP* of TSC1, together with HER3 and TAF1Lmissense mutations. Grade 3-4 adverse events (AE) were: 54.5% alopecia, 41% mucositis, 36.4% fatigue, 18% neutropenia (14% febrile neutropenia). There were 2 treatment-related deaths (septic shock). Conclusions: Preliminary data suggest the possibility for sustained disease control in about 20% of pts, including very long term PR/SD. Hb ≥ 10 and mutations of TSC1/mTOR (TAF1L?) might be the clinical and biological markers for patient enrichment design with SC inhibitors like Alisertib, although the incidence of AE is a major concern. Clinical trial information: NCT01653028.

479 Background: Little is known about the outcomes and prognostic factors of adjuvant chemotherapy for locally advanced PSCC after regional lymphadenectomy (LAD). Perioperative TPF combination regimen showed activity in PSCC in our series (Nicolai N et al, Clin Genitourin Cancer 2015). Methods: We retrospectively reviewed the clinical and histological data of 21 pts who had received 3-4 cycles TPF (q3wks) in the adjuvant setting at our center. Univariable and multivariable Cox regression analyses were done evaluating potential prognostic factors of disease-free (DFS) and overall survival (OS) of TPF: number of prior LAD, time from diagnosis of nodal disease to start TPF, time from LAD to start TPF, pN3, pelvic, and bilateral nodal disease, and p53 immunohistochemical (IHC) expression in nodal metastases. Results: Pts had received TPF from 07/04 to 07/12, after inguinal (n = 6) or inguinal + pelvic LAD (n = 15) and the median follow up was 52 months. LAD was radical in all cases and none had received radiotherapy. After surgery, 13 (61.9%) had pelvic and 5 (23.8%) bilateral inguinal nodal metastases. Median time from LAD to TPF was 5.4 wks (IQR: 4.1-7.3) and median time from diagnosis of N disease to TPF was 8.3 wks (IQR: 4.1-15.4). 11 of 19 available pts (57.9%) had p53+ve IHC. Univariably, only p53+ve IHC trended to poorer DFS (HR: 4.14, 95%CI: 0.87-19.68, p = 0.074) and OS (HR: 4.54, 95%CI: 0.95-21.56, p = 0.056). The same results were obtained multivariably for DFS (HR: 3.76, 95%CI: 0.78-17.96, p = 0.096) and OS (HR: 4.29, 95%CI: 0.89-20.57, p = 0.067). The bias-corrected c-indexes of the models were 0.64 and 0.62, respectively. Median DFS was 8.9 months (IQR: 5.9-22.7) for p53+ve pts vs not estimable (NE) for p53-ve pts (p = 0.051) and median OS was 17.2 months (IQR: 12.8-22.7) and NE, respectively (p = 0.037). Conclusions: In pts who had received adjuvant TPF for node-positive PSCC, p53 IHC positivity seemed to be associated with poorer outcome, and further study is warranted in larger datasets to confirm these findings. This information might be useful to select the optimal candidates to multimodal therapy for advanced PSCC.

BACKGROUND. The objective of this study was to investigate prognostic factors and clinical outcome of myxoid/round cell and pleomorphic liposarcoma. METHODS. Three hundred twenty-nine patients with localized myxoid/round cell or pleomorphic liposarcoma who underwent surgery at the Istituto Nazionale per lo Studio e la Cura dei Tumori (Milan, Italy) over 25 years were reviewed. The reates of local recurrence, distant metastases, and survival were studied. RESULTS. Two hundred fourteen patients presented with primary disease, and 115 patients had locally recurrent tumors. The disease-specific survival rate was 75% at 10 years, and the local recurrence and distant metastases incidence were 25% and 15%, respectively. Presentation with recurrent disease, tumor size (>10 cm), tumor grade (French Federation of Cancer Centers grade II or III vs grade I), and positive surgical margins were independent predictors of death. Tumor site and radiation therapy also played a role, mostly related to their effect on local outcome. Pathologic grade and histologic subtype influenced distant metastases. Extrapulmonary metastases were associated with poorer postmetastatic disease-specific survival. CONCLUSIONS. Myxoid/round cell liposarcomas shared similar prognostic factors with other soft tissue sarcomas and had a relatively good clinical outcome. The presence of >5% of round cell component singled out a group of patients at greater risk of metastases and death but with a broad spectrum of disease aggressiveness. Extrapulmonary metastases were a peculiar pattern of myxoid/round cell liposarcoma that require special consideration for treatment and prognosis. Cancer 2007. © 2007 American Cancer Society.

e15572 Background: About one third of patients with metastatic GCT are classified as IRGCT, and available guidelines recommend the same treatment of poor risk cases. Yet the prognosis of these pati...

e15559 Background: The optimal management of relapsed seminoma is still unknown. We retrospectively analyzed the outcomes of CDCT versus HDCT in the salvage treatment strategy of patients (pts) fai...

380 Background: IRGCT comprises a consistent category of metastatic patients (pts), and information on the recommended management of these pts should be updated. Usually they enter clinical trials for poor prognosis GCT. We aimed to address the heterogeneity of this category and to identify clinical prognostic factors for sub-stratification of pts. Methods: Data on consecutive pts with IRGCT and who received treatment at Fondazione INT Milano in the time-frame 02/1980-03/2014 were collected. Cox regression analyses were done evaluating potential prognostic factors for overall survival (OS, primary endpoint) to first-line therapy. Each factor was evaluated in a multivariable model. An exploratory OS comparison between outlier groups was undertaken with Kaplan Meier curves and logrank test. Results: Data on 181 pts were collected. Median age was 27 yrs (IQR 22-32), 10 pts had a retroperitoneal (RP) primary, 6 had pure seminoma. 72 (39.8%) had lung metastases and 54 (32.3%) bulky (i.e. ≥10cm) RP lymph-nodes (LN). Pts received cisplatin, bleomycin and etoposide (PEB, n=156) or vinblastine (PVB, n=23), 2 other treatments. Median follow up was 173 months (IQR: 87-237). Globally, 5-y PFS and OS were 66.8% (95%CI: 60.1-74.2) and 83.3% (77.8-89.2). However, 5-y OS for pts with AFP 5,000-10,000 IU/ml (N=19) was 61.8% (95%CI: 43.0-88.7) while it was 89.1% (95%CI: 81.2-97.7) for nonseminomas with elevated LDH only (N=57) and similar for elevated HCG only (N=22); overall p

Purpose The selection of effective schedules of treatment for metastatic non-small cell lung cancer still remains a challenge for the oncologist. The present multicentric phase II study was designed in order to investigate the activity and safety of the combination of weekly paclitaxel and celecoxib as second-line treatment for non-small cell lung cancer. As a secondary endpoint, the possible correlation of biomarkers with objective response was investigated in a subset of patients. Patients and methods Patients with platinum-refractory non-small cell lung cancer and Eastern Cooperative Oncology Group performance status 0-2 entered the present phase II study. Paclitaxel was administered at the dose of 80 mg/m(2) i.v. weekly for 6 weeks, followed by a 2-week rest, and celecoxib, 400 mg p.o. b.i.d. administered continuously. A cycle consisted of 8 weeks of treatment. Determination of circulating vascular endothelial growth factor and interleukin 6 was performed at baseline and every two cycles. Results Fifty-eight patients were enrolled: median age, 60 years (range, 30-77 years); male/female ratio = 44/14; performance status, 0, 31 patients; 1, 25 patients; and 2, two patients. Predominant histotype was adenocarcinoma (34 cases), and most patients had at least two sites of disease. According to the intent-to-treat analysis, 14/58 objective responses (24.1%) and 24/58 (41.3%) stabilizations of disease were observed, with a median duration of 4 months (range, 2-22+ months) and 5 months (range, 1-13 months), respectively. Median time to progression and median overall survival were 5 and 11 months, respectively. One-year survival was 42.5%. The main toxicity was neuropathy (4% of grade 3). Preliminary results suggest that decrease in serum vascular endothelial growth factor level is significantly associated with clinical response. Discussion Combination of celecoxib and weekly paclitaxel is safe and active new regimen in pretreated non-small cell lung cancer. Toxicity appears not to be worsened by the addition of celecoxib. According to preliminary results, serum vascular endothelial growth factor level seems to be predictive of response, suggesting that it should be further investigated as a surrogate marker of response.

Background. The present study was designed with the aim of evaluating the tolerability and activity of pegylated liposomial doxorubicin (PLD) in combination with weekly docetaxel as first line treatment of advanced breast cancer. Patients and methods. Fifty-seven patients entered the study. PLD was administered at escalating doses starting from 30 mg/m, on day 1; docetaxel was administered at the fixed dose of 35 mg/m on days 2 and 9. A cycle of therapy consisted of 21 days. Results. The MTD was achieved at the dose of 40 mg/m of PLD, being febrile neutropenia and palmar-plantar-erythrodisesthesia (PPE) the dose-limiting toxicities (DLTs), so that the fixed dose of PLD for the Phase II study was 35 mg/m. Forty-two consecutive patients received treatment at the established dose for a total of 194 cycles: among these, three patients were withdrawn for severe allergic reaction at the first administration of PLD. Hematological toxicity was moderate, the most common grade 1–3 non-hematological toxicities were stomatitis and PPE, occurring in 20 (47.5%) and 16 (38%) patients, respectively. No cardiac toxicity was recorded. According to the intent to treat analysis a major objective response was observed in 59.5% of patients (95% CI, 43.3–74.4%), with a median time to progression of 9months and an estimated overall survival at 18 months of 62%. Conclusion. The combination of PLD and weekly docetaxel is an effective first-line therapy for patients with advanced breast cancer. PPE and mucositis are the most relevant side effects of such a combination.

BACKGROUND The objective of this study was to assess long-term prognosis and patterns of failure in patients with retroperitoneal soft tissue sarcoma who underwent surgery with curative intent at a single institution. Several series of patients with retroperitoneal sarcoma were reported, providing widely variable data regarding local and distant failure. METHODS Overall, 167 consecutive patients who underwent surgery for retroperitoneal soft tissue sarcoma over a 20 year span at a single referral center were reviewed retrospectively. Eighty-two patients presented with primary disease, whereas 85 patients had recurrent locoregional tumors. Surgical resection was considered macroscopically complete in 147 of 167 patients (88%). RESULTS Overall survival at 10 years after definitive surgery was approximately 27%, and the disease-free survival was approximately 16%. Only a minority of patients developed distant metastases. The risk of recurrence was correlated with whether patients underwent surgery for primary disease or for recurrent disease; the 10 year disease-free survival rate was 27% in the former group and 4.6% in the latter group. Histotype and malignancy grade were other prognostic factors, with the former found to be predictive of the pattern of failure and the latter predictive of overall survival. CONCLUSIONS Local recurrence after primary surgery and high-grade malignancy were associated with the worst survival. Histologic subtype appeared to influence the pattern of recurrence, which mainly was local for patients with liposarcoma but was both local and distant for patients with the other histotypes. Cancer 2004. © 2004 American Cancer Society.

Purpose: Prognosis of patients with advanced oral cavity cancer is worth improving. Chemotherapy has been reported to be especially active in oral cavity tumors. Here we repeat the results of a randomized, multicenter trial enrolling patients with a resectable, stage T2–T4 (> 3 cm), N0–N2, M0 untreated, squamous cell carcinoma of the oral cavity. Patients and Methods: Patients were randomly assigned to three cycles of cisplatin and fluorouracil followed by surgery (chemotherapy arm) or surgery alone (control arm). In both arms, postoperative radiotherapy was reserved to high-risk patients, and surgery was modulated depending on the tumor’s closeness to the mandible. Patients’ accrual was opened in 1989 and closed in 1999. It included 195 patients. Results: In the chemotherapy arm, three toxic deaths were recorded. No significant difference in overall survival was found. Five-year overall survival was, for both arms, 55%. Postoperative radiotherapy was administered in 33% of patients in the chemotherapy arm, versus 46% in the control arm. A mandible resection was performed in 52% of patients in the control arm, versus 31% in the chemotherapy arm. Conclusion: The addition of primary chemotherapy to standard surgery was unable to improve survival. However, in this study, primary chemotherapy seemed to play a role in reducing the number of patients who needed to undergo mandibulectomy and/or radiation therapy. Variations in the criteria used to select patients for these treatment options may make it difficult to generalize these results, but there appears to be room for using preoperative chemotherapy to spare demolitive surgery and/or radiation therapy in patients with advanced, resectable oral cavity cancer.

4547 Background: Knowledge of the expression of molecular drivers and potentially druggable targets may enhance prognostic classification of metastatic UC. We aimed at assessing the expression of m...

373 Background: Mediastinal GCTs and PMNSGCTs poorly benefit from CT and half of patients (pts) still die for disease. Enhancing the risk stratification may result in tailoring a personalized treatment strategy since diagnosis. Methods: Between 1985 and 2012, 87 pts with PM-GCT were treated at our center. Of them, pure seminomatous histology was excluded. Cox proportional hazards regression analysis was conducted to examine the prognostic impact of these candidate factors on overall survival (OS): type of 1st-line CT (high [HDCT] vs conventional dose [CDCT]), post-CT surgery, type of baseline elevated serum tumor marker (STM), presence of lung or liver-bone-brain metastases (LBB), STM response (still elevated vs normal or normalized), and histology (viable cancer [VC] vs necrosis/teratoma [NT]). OS curves were compared by Kaplan Meier method with the log-rank test. Results: The study included 68 cases with PMNSGCT. Median age was 28.5 yrs (IQR: 23-35). 12 pts (17.7%) presented with mediastinal syndrome, 23 (33.8%) had lung and 7 (10.3%) LBB metastases. 12 pts received upfront HDCT and 45 pts (66.2%) underwent post-CT surgery. The final model of poor prognostic factors included no surgery (HR: 8.74, 95%CI, 1.77-43.01), surgery + VC (HR: 6.97, 95%CI, 1.46-33.30), and lung metastases (HR: 2.92, 95%CI, 0.99-8.64). The model demonstrated moderate discriminatory ability for OS (c-statistic=0.68). A risk stratification model based on the combination of these factors and accounting for a 50% five-year survival cutoff identified 2 groups (poor prognosis, N=28 vs good prognosis, N=26) with distinct overall survival curves (p