Refine
Document Type
- Article (2)
- Conference Proceeding (2)
- Part of a Book (1)
Has Fulltext
- yes (5)
Is part of the Bibliography
- no (5)
Keywords
- Computerlinguistik (2)
- Deutsch (1)
- Generic NLP Architecture (1)
- HPSG Parsing (1)
- IE (1)
- Literarische Romantik / Deutsch (1)
- Literatur (1)
- Maschinelle Übersetzung (1)
- Romantik (1)
- Shallow NLP (1)
Institute
- Extern (2)
- Georg-Speyer-Haus (2)
- Medizin (2)
In this paper we show an approach to the customization of GermaNet to the German HPSG grammar lexicon developed in the Verbmobil project. GermaNet has a broad coverage of the German base vocabulary and fine-grained semantic classification; while the HPSG grammar lexicon is comparatively small und has a coarse-grained semantic classification. In our approach, we have developed a mapping algorithm to relate the synsets in GermaNet with the semantic sorts in HPSG. The evaluation result shows that this approach is useful for the lexical extension of our deep grammar development to cope with real-world text understanding.
Die deutsche Romantik ist eine Zeit des Aufbruchs, der kühnen und spielerischen Anfangsexperimente; aber sie ist auch eine Zeit der Krise. Zwar herrscht sowohl in der literaturwissenschaftlichen Forschung wie in den weiter gespannten Versuchen einer kulturellen Diagnostik Einverständnis darüber, daß "Modernen" sich in wiederholten Schüben ereignet haben [...] und daß jede dieser neu erscheinenden Modernen an den Errungenschaften der vorangehenden partizipiert. Doch könnte man behaupten, daß die Krise, die sich in der deutschen Romantik abzeichnet, zu den
differenziertesten gehört, und daß sie in den Umbrüchen, die aus ihr resultierten, alle Felder des Wissens, des Imaginierens und des Darstellens nachhaltig affizierte und transformierte.
Introduction: ScFv(FRP5)-ETA is a recombinant antibody toxin with binding specificity for ErbB2 (HER2). It consists of an N-terminal single-chain antibody fragment (scFv), genetically linked to truncated Pseudomonas exotoxin A (ETA). Potent antitumoral activity of scFv(FRP5)-ETA against ErbB2-overexpressing tumor cells was previously demonstrated in vitro and in animal models. Here we report the first systemic application of scFv(FRP5)-ETA in human cancer patients.
Methods: We have performed a phase I dose-finding study, with the objective to assess the maximum tolerated dose and the dose-limiting toxicity of intravenously injected scFv(FRP5)-ETA. Eighteen patients suffering from ErbB2-expressing metastatic breast cancers, prostate cancers, head and neck cancer, non small cell lung cancer, or transitional cell carcinoma were treated. Dose levels of 2, 4, 10, 12.5, and 20 μg/kg scFv(FRP5)-ETA were administered as five daily infusions each for two consecutive weeks.
Results: No hematologic, renal, and/or cardiovascular toxicities were noted in any of the patients treated. However, transient elevation of liver enzymes was observed, and considered dose limiting, in one of six patients at the maximum tolerated dose of 12.5 μg/kg, and in two of three patients at 20 μg/kg. Fifteen minutes after injection, peak concentrations of more than 100 ng/ml scFv(FRP5)-ETA were obtained at a dose of 10 μg/kg, indicating that predicted therapeutic levels of the recombinant protein can be applied without inducing toxic side effects. Induction of antibodies against scFv(FRP5)-ETA was observed 8 days after initiation of therapy in 13 patients investigated, but only in five of these patients could neutralizing activity be detected. Two patients showed stable disease and in three patients clinical signs of activity in terms of signs and symptoms were observed (all treated at doses ≥ 10 μg/kg). Disease progression occurred in 11 of the patients.
Conclusion: Our results demonstrate that systemic therapy with scFv(FRP5)-ETA can be safely administered up to a maximum tolerated dose of 12.5 μg/kg in patients with ErbB2-expressing tumors, justifying further clinical development.
Following publication of the data presented by von Minckwitz and colleagues it has been brought to our attention that some patients should be scored differently. Stable disease was seen in three of the eighteen patients instead of two of the eighteen patients: one patient with transitional cell carcinoma treated at 4 µg/kg scFv(FRP5)-ETA per day, and two breast cancer patients treated at 4 and 12.5 µg/kg scFv(FRP5)-ETA per day. Disease progression occured in 9 of the eighteen patients evaluated (see corrected Table 2 overleaf). This does not affect the conclusions of our study. In addition we would like to correct the following errors: patient IDs for patients U01 and U02 in the original Table 2 were interchanged. In addition, patient N03 had a grade 3 elevation of gamma-glutamyl transferase, and not grade 2 (see corrected Table 2 overleaf).
We present an architecture for the integration of shallow and deep NLP components which is aimed at flexible combination of different language technologies for a range of practical current and future applications. In particular, we describe the integration of a high-level HPSG parsing system with different high-performance shallow components, ranging from named entity recognition to chunk parsing and shallow clause recognition. The NLP components enrich a representation of natural language text with layers of new XML meta-information using a single shared data structure, called the text chart. We describe details of the integration methods, and show how information extraction and language checking applications for realworld German text benefit from a deep grammatical analysis.