Abstract P2-08-47: Evaluation of stromal tumor-infiltrating lymphocytes (sTIL) and tertiary lymphoid structures (TLS) in early breast cancer patients with triple negative breast cancer(TNBC) included in a prospective study of neoadjuvant chemotherapy (NAC) with Epirubicin and cyclophosphamide (EC) and carboplatin-paclitaxel (PC) (BSMO 2014-01)

Background: BSMO 2014-01 is a completed prospective phase 2 study evaluating the efficacy of neoadjuvant EC and PC. One of the secondary endpoints was the correlation of sTIL with response, pCR and survival. We also assessed the relationship between sTIL and TLS in the diagnostic biopsies.

Methods: Stromal TIL (sTIL) were evaluated on H&E stained tumor biopsies before the start of the NAC according to the criteria described by Salgado et al(1). Scores were defined as "low" or "high" if lymphocytic infiltration in the stroma around the tumor was ≤ 10% or > 10%. TLS are ectopic lymph node-like structures recently identified in breast cancer. TLS were counted using a dual IHC stain for CD3 (T cells) and CD20 (B cells) and categorized as "little" if the TLS occupied < 10% or "moderate to abundant" if they occupied ≥ 10% of the adjacent tissue. The correlation between sTIL and pathologic parameters was analyzed using the chi-square test; DFS and OS between the groups was estimated by using the log-rank test.

Results: So far we could quantify the number of sTIL in 38 out of 63 TNBC pts treated with neoadjuvant EC-PC. Twenty eight pts had a high sTIL score and 10 pts had a low sTIL score. The high-sTIL group (19/28) achieved a numerical higher pathologic complete remission (pCR) rate than the low-sTIL group (5/10) (p=0.3); both groups had a comparable disease free survival of 28.6 mths and 26.7 mths respectively (p=0.7). The overall survival was similar:29 mths and 27.8 mths respectively (p=0.8). Stromal TLS were identified in 10 out 23 samples available for this analysis and we could demonstrate a positive correlation between high levels of sTIL and high levels of moderate to abundant TLS(CD3) in the adjacent tissue in six out of the ten samples in which TLS were present (p=0.1).

Conclusion: These preliminary results could not confirm the results published by Denkert et al earlier this year(2). A trend for correlation of the presence of high sTIL with moderate to abundant levels of TLS was found. Analysis on the remaining samples of all patients included in the study and correlation with outcome is ongoing and these completed results will be presented.

(1)Assessing Tumor-infiltrating lymphocytes in solid tumors. Hendry S, Salgado R et al. Adv Anat Pathol 2017; 235-251.

(2)Tumour-infiltrating lymphocytes and prognosis in different subtypes of breast cancer. Denkert C et al. Lancet Oncol 2018; 40-50.

Citation Format: Fontaine C, Van den Eynden G, de Wind R, Boisson A, Renard V, Van den Bulck H, Vuylsteke P, Glorieux P, Dopchie C, Decoster L, De Grève J, Awada A, Wildiers H, Willard-Gallo K. Evaluation of stromal tumor-infiltrating lymphocytes (sTIL) and tertiary lymphoid structures (TLS) in early breast cancer patients with triple negative breast cancer(TNBC) included in a prospective study of neoadjuvant chemotherapy (NAC) with Epirubicin and cyclophosphamide (EC) and carboplatin-paclitaxel (PC) (BSMO 2014-01) [abstract]. In: Proceedings of the 2018 San Antonio Breast Cancer Symposium; 2018 Dec 4-8; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2019;79(4 Suppl):Abstract nr P2-08-47.

Recurrent invasive thymoma with pleural dissemination : disease management and treatment possibilities

Thymoma is the most common benign neoplasm of the anterior mediastinum presenting often an agressive behaviour typical for the malignants tumors. The rate of invasive thymoma recurrency is relatively high. We present the case of a 55-year old man with a recurrent invasive thymoma with a pleural dissemination, detected on CT-imaging 2 years following his primary surgery. Since the first pre-operative imaging studies showed no invasion of the adjacent organs and a thymoma was suspected, a surgical resection was decided as a first line treatment. Per-operatively a number of adjacent structures were invaded and despite a macroscopical RO resection, the margins were microscopically positive. An invasive thymoma, WHO classification B3, Masaoka stage IVb was diagnosed and the patient received adjuvant radiotherapy. We highlight the role of multimodality treatement and disscus the potential of surgical, radiotherapeutical and systemic therapy in stage IV thymoma as well as in recurrent disease.

[Mastocytosis: revisited with new cytogenetic data]

Mastocytosis is a heterogenous disorder due to abnormal proliferation and infiltration of mast cells in different tissues, primarily the skin and the bone marrow. Cutaneous mastocytosis is often benign and regresses spontaneously. Systemic mastocytosis is a chronic disease in which some types are indolent but other types such as mast cell leukemia are very aggressive. Pathogenesis of systemic mastocytosis involves a somatic mutation of the gene coding for the c-kit receptor, the most frequent mutation being D816V. Diagnostic criteria have been established by the WHO using histopathological, molecular and biochemical parameters. Treatment of systemic mastocytosis remains a challenge for the clinician due to variability and complexity of the disease. There is, in addition, a lack of a standard and efficient treatment. New targeted therapies with tyrosine kinase inhibitors directed against the c-kit receptor are currently being studied, with the purpose to act specifically on the " primum movens "of the disease. The current review provides an overview of pathogenesis, clinical presentation, diagnosis and classification of cutaneous and systemic mastocytosis. We also discuss the prognosis and the different treatments currently available according to the sub-type of mastocytosis.

High-yield secretion of recombinant gelatins by Pichia pastoris

Recombinant non-hydroxylated gelatins based on mouse type I and rat type III collagen sequences were secreted from the methylotrophic yeast Pichia pastoris, using the Saccharomyces cerevisiae alpha-mating factor prepro signal. Proteolytic degradation could be minimized to a large extent by performing fermentations at pH 3.0 and by adding casamino acids to the medium, even though gelatin is extremely susceptible to proteolysis due to its open, unfolded structure. Proteolytic cleavage at specific mono-arginylic sites, by a putative Kex2-like protease, could be successfully abolished by site-directed mutagenesis of these sites. Production levels as high as 14.8 g/l clarified both were obtained, using multicopy tranformants. To our knowledge, this represents the highest level of heterologous protein secretion reported to date for P. pastoris.

High-yield secretion of recombinant gelatins by Pichia pastoris

Recombinant non-hydroxylated gelatins based on mouse type I and rat type III collagen sequences were secreted from the methylotrophic yeast Pichia pastoris, using the Saccharomyces cerevisiae alpha-mating factor prepro signal. Proteolytic degradation could be minimized to a large extent by performing fermentations at pH 3.0 and by adding casamino acids to the medium, even though gelatin is extremely susceptible to proteolysis due to its open, unfolded structure. Proteolytic cleavage at specific mono-arginylic sites, by a putative Kex2-like protease, could be successfully abolished by site-directed mutagenesis of these sites. Production levels as high as 14.8 g/l clarified both were obtained, using multicopy tranformants. To our knowledge, this represents the highest level of heterologous protein secretion reported to date for P. pastoris.

Engineering of factors determining alpha-amylase and cyclodextrin glycosyltransferase specificity in the cyclodextrin glycosyltransferase from Thermoanaerobacterium thermosulfurigenes EM1

The starch-degrading enzymes alpha-amylase and cyclodextrin glycosyltransferase (CGTase) are functionally and structurally closely related, with CGTases containing two additional domains (called D and E) compared to the three domains of alpha-amylases (A, B and C). Amino acid residue 196 (Thermoanaerobacterium thermosulfurigenes EM1 CGTase numbering) occupies a dominant position in the active-site cleft. All alpha-amylases studied have a small residue at this position (Gly, Leu, Ser, Thr or Val), in contrast to CGTases which have a more bulky aromatic residue (Tyr or Phe) at this position, which is highly conserved. Characterization of the F196G mutant CGTase of T. thermosulfurigenes EM1 revealed that, for unknown reasons, apart from the F196G mutation, domain E as well as a part of domain D had become deleted [mutant F196G(delta'DE)]. This, nevertheless, did not prevent the purification of a stable and active mutant CGTase protein (62 kDa). The mutant protein was more similar to an alpha-amylase protein in terms of the identity of residue 196, and in the domain structure containing, however, some additional C-terminal structure. The mutant showed a strongly reduced temperature optimum. Due to a frameshift mutation in mutant F196G, a separate protein of 19 kDa with the DE domains was also produced. Mutant F196G(delta'DE) displayed a strongly reduced raw-starch-binding capacity, similar to the situation in most alpha-amylases that lack a raw-starch-binding E domain. Compared to wild-type CGTase, cyclization, coupling and disproportionation activities had become drastically reduced in the mutant F196G(delta'DE), but its saccharifying activity had doubled, reaching the highest level ever reported for a CGTase. Under industrial production process conditions, wild-type CGTase converted starch into 35% cyclodextrins and 11% linear oligosaccharides (glucose, maltose and maltotriose), whereas mutant F196G(delta'DE) converted starch into 21% cyclodextrins and 18% into linear oligosaccharides. These biochemical characteristics indicate a clear shift from CGTase to alpha-amylase specificity.

Engineering of cyclodextrin product specificity and pH optima of the thermostable cyclodextrin glycosyltransferase from Thermoanaerobacterium thermosulfurigenes EM1

The product specificity and pH optimum of the thermostable cyclodextrin glycosyltransferase (CGTase) from Thermoanaerobacterium thermosulfurigenes EM1 was engineered using a combination of x-ray crystallography and site-directed mutagenesis. Previously, a crystal soaking experiment with the Bacillus circulans strain 251 beta-CGTase had revealed a maltononaose inhibitor bound to the enzyme in an extended conformation. An identical experiment with the CGTase from T. thermosulfurigenes EM1 resulted in a 2.6-A resolution x-ray structure of a complex with a maltohexaose inhibitor, bound in a different conformation. We hypothesize that the new maltohexaose conformation is related to the enhanced alpha-cyclodextrin production of the CGTase. The detailed structural information subsequently allowed engineering of the cyclodextrin product specificity of the CGTase from T. thermosulfurigenes EM1 by site-directed mutagenesis. Mutation D371R was aimed at hindering the maltohexaose conformation and resulted in enhanced production of larger size cyclodextrins (beta- and gamma-CD). Mutation D197H was aimed at stabilization of the new maltohexaose conformation and resulted in increased production of alpha-CD. Glu258 is involved in catalysis in CGTases as well as alpha-amylases, and is the proton donor in the first step of the cyclization reaction. Amino acids close to Glu258 in the CGTase from T. thermosulfurigenes EM1 were changed. Phe284 was replaced by Lys and Asn327 by Asp. The mutants showed changes in both the high and low pH slopes of the optimum curve for cyclization and hydrolysis when compared with the wild-type enzyme. This suggests that the pH optimum curve of CGTase is determined only by residue Glu258.

Crystal structure at 2.3 A resolution and revised nucleotide sequence of the thermostable cyclodextrin glycosyltransferase from Thermonanaerobacterium thermosulfurigenes EM1

The crystal structure of the cyclodextrin glycosyltransferase (CGTase) from the thermophilic microorganism Thermoanaerobacterium thermosulfurigenes EM1 has been elucidated at 2.3 A resolution. The final model consists of all 683 amino acid residues, two calcium ions and 343 water molecules, and has a crystallographic R-factor of 17.9% (Rfree 24.9%) with excellent stereochemistry. The overall fold of the enzyme is highly similar to that reported for mesophilic CGTases and differences are observed only at surface loop regions. Closer inspection of these loop regions and comparison with other CGTase structures reveals that especially loops 88-95, 335-339 and 534-539 possibly contribute with novel hydrogen bonds and apolar contacts to the stabilization of the enzyme. Other structural features that might confer thermostability to the T. thermosulfurigenes EM1 CGTase are the introduction of five new salt-bridges and three Gly to Ala/Pro substitutions. The abundance of Ser, Thr and Tyr residues near the active site and oligosaccharide binding sites might explain the increased thermostability of CGTase in the presence of starch, by allowing amylose chains to bind non-specifically to the protein. Additional stabilization of the A/E domain interface through apolar contacts involves residues Phe273 and Tyr187. No additional or improved calcium binding is observed in the structure, suggesting that the observed stabilization in the presence of calcium ions is caused by the reduced exchange of calcium from the protein to the solvent, rendering it less susceptible to unfolding. The 50% decrease in cyclization activity of the T. thermosulfurigenes EM1 CGTase compared with that of B. circulans strain 251 appears to be caused by the changes in the conformation and amino acid composition of the 88-95 loop. In the T. thermosulfurigenes EM1 CGTase there is no residue homologous to Tyr89, which was observed to take part in stacking interactions with bound substrate in the case of the B. circulans strain 251 CGTase. The lack of this interaction in the enzyme-substrate complex is expected to destabilize bound substrates prior to cyclization. Apparently, some catalytic functionality of CGTase has been sacrificed for the sake of structural stability by modifying loop regions near the active site.

Cyclodextrin formation by the thermostable alpha-amylase of Thermoanaerobacterium thermosulfurigenes EM1 and reclassification of the enzyme as a cyclodextrin glycosyltransferase

Extensive characterization of the thermostable alpha-amylase of Clostridium thermosulfurogenes EM1, recently reclassified as Thermoanaerobacterium thermosulfurigenes, clearly demonstrated that the enzyme is a cyclodextrin glycosyltransferase (CGTase). Product analysis after incubation of the enzyme with starch revealed formation of alpha-, beta-, and gamma-cyclodextrins, as well as linear sugars. The specific activity for cyclization of this CGTase was similar to those of other CGTases, whereas the specific activity for hydrolysis was relatively high in comparison with other CGTases. Alignment of the amino acid sequence of the T. thermosulfurigenes enzyme with sequences from known bacterial CGTases showed high homology. The four consensus regions of carbohydrate-converting enzymes, as well as a C-terminal raw-starch binding motif, could be identified in the sequence.

Kinetics of Bacterial Growth on Chlorinated Aliphatic Compounds

With the pure bacterial cultures Ancylobacter aquaticus AD20 and AD25, Xanthobacter autotrophicus GJ10, and Pseudomonas sp. strain AD1, Monod kinetics was observed during growth in chemostat cultures on 1,2-dichloroethane (AD20, AD25, and GJ10), 2-chloroethanol (AD20 and GJ10), and 1,3-dichloro-2-propanol (AD1). Both the Michaelis-Menten constants ( K m ) of the first catabolic (dehalogenating) enzyme and the Monod half-saturation constants ( K s ) followed the order 2-chloroethanol, 1,3-dichloro-2-propanol, epichlorohydrin, and 1,2-dichloroethane. The K s values of strains GJ10, AD20, and AD25 for 1,2-dichloroethane were 260, 222, and 24 μM, respectively. The low K s value of strain AD25 was correlated with a higher haloalkane dehalogenase content of this bacterium. The growth rates of strains AD20 and GJ10 in continuous cultures on 1,2-dichloroethane were higher than the rates predicted from the kinetics of the haloalkane dehalogenase and the concentration of the enzyme in the cells. The results indicate that the efficiency of chlorinated compound removal is indeed influenced by the kinetic properties and cellular content of the first catabolic enzyme. The cell envelope did not seem to act as a barrier for permeation of 1,2-dichloroethane.