Abstract P4-02-10: Are fine-needle aspiration (FNA)-derived cell blocks a useful tissue sample surrogate for testing conventional biomarkers and PD-L1 in breast cancer?

Background The diagnosis of breast cancer (BC) is based on clinical examination in combination with imaging and confirmed by pathological assessment of core needle biopsy (CNB) or fine needle aspiration (FNA). The biological profile of the lesion is needed to define prognosis and guide therapy. Given the importance of an early and minimally invasive diagnosis, we aimed to verify whether the biological features detected in FNA-derived cytological material reflect the biological characteristics of surgical specimens.

Methods We used immunohistochemistry and fluorescence in situ hybridization (FISH) to study a panel of biomarkers (ER, PgR, Ki67 and HER2 in 93 patients, programmed death-ligand 1 (PD-L1) in 20 patients) in FNA-derived cell blocks of BC, comparing the results with those obtained on the histological specimens.Immunostaining was performed with the Ventana Benchmark XT system and the Ultraview DAB Detection Kit (Ventana Medical Systems). Confirm anti-ER (clone SP1, Ventana), Confirm Anti-PgR (clone 1E2, Ventana) and Ki67 (clone Mib-1, Dako, Carpinteria, CA, US) antibodies were used. Ventana PD-L1 (SP263) assay (Ventana Medical Systems) was used for PD-L1 immunostaining. HER2 status was analyzed by FISH using PathVysion kit (Abbott Molecular, Abbott Park, Illinois, IL, USA).

Results Median immunopositive values of ER, PgR Ki67, and PD-L1 were similar in cell blocks and surgical samples. Concordance for ER and PgR between FNA and histological samples was 98% and 84%, respectively. With regard to Ki67 and HER2 status, concordance between the two specimen types was 90% and 96%, respectively. PD-L1 expression analyzed in FNA-derived samples was 100% concordant with that of surgical specimens. Tumor subtype classification for triple-negative and HER2-positive tumors in FNA samples was always concordant with the subtype determined in surgical material.

Table 1.Concordance of tumor subtype classification between FNAB-derived and histological samples.Overall series (n=93)87%Luminal A (n=26)81%Luminal B (n=47)86%HER2-positive (n=6)100%Triple-negative (n=14)100%

Conclusions We showed that biological marker determination in FNA-derived cell blocks is feasible and provides useful information and comparable results with those obtained by histological evaluation. Given the low cost of the procedure and its minimal impact on patients, that cytological samples could be used as an alternative to tissue samples for early BC biomarker evaluation to facilitate the planning of tailored neoadjuvant therapy.

Citation Format: Bravaccini S, Ravaioli S, Rocca A, Tumedei MM, Bucchi E, Malmesi M, Medri L, Bonafè M, Puccetti M. Are fine-needle aspiration (FNA)-derived cell blocks a useful tissue sample surrogate for testing conventional biomarkers and PD-L1 in breast cancer? [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 P4-02-10.

Abstract P2-04-01: Human papillomavirus (HPV) DNA detection in breast cancer by liquid biopsy: Something new on the horizon?

Background The presence of human papillomavirus (HPV) DNA in breast cancer (BC) tissues has been widely investigated in recent years. HPV DNA has been detected in ductal lavage fluids and in serum-derived extracellular vesicles of patients with benign or in situ breast lesions. However, there are no data attesting to its presence in liquid biopsies of different BC subtypes or to its impact on prognosis.

Methods We analyzed a total of 72 serum samples for the presence of circulating HPV DNA, of which 20 were from luminal A BC (5 relapsed, 15 non relapsed), 17 from luminal B BC (5 relapsed, 12 non relapsed), 15 from triple-negative BC (6 relapsed and 9 non relapsed), 12 from HER2-positive BC (3 relapsed, 9 non relapsed) and 8 from healthy subjects. Circulating DNA was extracted and purified from 500 μl of serum by Qiamp DNA minikit (Qiagen, Milan, Italy) according to the manufacturer's instructions. HPV DNA was assessed by a high-throughput MALDI-TOF mass spectrometry-based method (Mass Array Platform, Agena Bioscience, Hamburg, Germany). The DNA target sequence was amplified by a multiplex PCR with HPV E6 or E7 gene-specific primers. A primer for primer extension annealing to the amplified product was extended at its 3′ terminal base for each HPV type.

Results HPV DNA was detected in 5 BC patients but in none of the healthy controls. None of the serum samples analyzed showed HPV DNA types 16 or 18. Four of the 5 BC cases had high-risk HPV DNA (type 39,45,52,59) and one had low-risk HPV DNA (type 73). The 4 patients with high risk HPV DNA had low-grade cervical intraepithelial neoplasia (CIN 1) detected by Pap smear prior to the diagnosis of BC. No relation was found between HPV infection and tumor subtype or prognosis. Our in vitro studies also revealed the active release of HPV DNA into the extracellular vesicle compartment of cervical cancer cells.

Table 1.Features of BC patients positive for circulating HPV DNA.Sample IDAge (years)GradingSubtypeDeathRelapseHPV type, riskCervical lesion17471Luminal A-No59, highCIN 139783Luminal B20/03/2015No45, highCIN 145462Luminal B-No39, highCIN 160583TN03/01/2011Yes73, lowHyperplasia63523HER2-positive-Yes52, highCIN 1TN, triple negative; CIN 1:low-grade cervical intraepithelial neoplasia

Conclusions Our findings support the feasibility of HPV DNA evaluation by liquid biopsy in BC. They also suggest that circulating HPV DNA in BC patients might be of cervical tissue origin and that the presence of HPV DNA in BC may be a consequence of its spreading from virus-infected tissue such as that of the uterine cervix.

Citation Format: Bravaccini S, Ravaioli S, Rocca A, Maltoni R, Cristalli C, Marasco E, De Carolis S, Cricca M, Bonafè M. Human papillomavirus (HPV) DNA detection in breast cancer by liquid biopsy: Something new on the horizon? [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 P2-04-01.

The selection of appropriate bacterial strains in preclinical evaluation of infection-resistant biomaterials

Implant-related infections are broadly recognized as one of the most serious and devastating complications associated with the use of biomaterials in medical practice. The growing interest and need for the development of implant materials with reduced susceptibility to microbial colonization and biofilm formation has necessitated the development of a series of in vitro and in vivo models for evaluation and preclinical testing. Current technologies provide these investigations with an ample choice of qualitative and quantitative techniques for an accurate assessment of the bioactivity and anti-infective efficacy of any new compound or device. These tests are typically performed using a reference bacterial strain designated as the test or reference strain. Recent molecular epidemiological studies have identified the complex clonal nature of most prevalent etiological agents implicated in implant-associated infections. New information which is continually emerging on the identity and the characteristics of both sporadic and epidemic clones must be considered when selecting a reference. A new emerging requirement is that the strain should be representative of the clones causing clinically relevant infections; they should, therefore, belong to the most prevalent epidemic clones rather than to sporadic ones, which may occur in only 1 out of 200 infections or even fewer. The correct choice of reference strain for preclinical tests is of crucial importance for the clinical significance of the achieved results. In this paper we report our experience and recommendations regarding this issue.

Prevalence of genes encoding for staphylococcal leukocidal toxins among clinical isolates of Staphylococcus aureus from implant orthopedic infections

Staphylococcus aureus has emerged as a major cause of implant infections. It is known that it is able to produce several toxins that contribute to its armory of virulent weapons, but there are still no data on their prevalence among isolates recovered from biomaterial-centered infections. In this study, 200 Staphylococcus aureus isolates from infections related to different types of orthopedic implants (hip and knee arthroprostheses, internal and external fixation devices) were tested by polymerase chain reaction for the prevalence of genes encoding for leukotoxins. Although almost all isolates were positive for the ã-hemolysin gene (99%), none was positive for lukM. The leukotoxin genes lukE/lukD were found in 67% of isolates. The presence of lukE/lukD was significantly associated with that of Accessory Gene Regulatory locus agr II. The lukE/lukD-positive isolates were significantly more prevalent in the staphylococcal isolates from knee arthroprostheses than in the isolates from the other implant types. The genes encoding Panton-Valentine leukocidin components were detected in only one isolate that, curiously enough, was taken solely from a knee arthroprosthesis infection.