Skip to main content
Download PDF
  • Research article
  • Open access
  • Published:

A five-gene reverse transcription-PCR assay for pre-operative classification of breast fibroepithelial lesions

Breast Cancer Research volume 18, Article number: 31 (2016) Cite this article

A Letter to this article was published on 25 July 2016

Abstract

Background

Breast fibroepithelial lesions are biphasic tumors and include fibroadenomas and phyllodes tumors. Preoperative distinction between fibroadenomas and phyllodes tumors is pivotal to clinical management. Fibroadenomas are clinically benign while phyllodes tumors are more unpredictable in biological behavior, with potential for recurrence. Differentiating the tumors may be challenging when they have overlapping clinical and histological features especially on core biopsies. Current molecular and immunohistochemical techniques have a limited role in the diagnosis of breast fibroepithelial lesions. We aimed to develop a practical molecular test to aid in distinguishing fibroadenomas from phyllodes tumors in the pre-operative setting.

Methods

We profiled the transcriptome of a training set of 48 formalin-fixed, paraffin-embedded fibroadenomas and phyllodes tumors and further designed 43 quantitative polymerase chain reaction (qPCR) assays to verify differentially expressed genes. Using machine learning to build predictive regression models, we selected a five-gene transcript set (ABCA8, APOD, CCL19, FN1, and PRAME) to discriminate between fibroadenomas and phyllodes tumors. We validated our assay in an independent cohort of 230 core biopsies obtained pre-operatively.

Results

Overall, the assay accurately classified 92.6 % of the samples (AUC = 0.948, 95 % CI 0.913–0.983, p = 2.51E-19), with a sensitivity of 82.9 % and specificity of 94.7 %.

Conclusions

We provide a robust assay for classifying breast fibroepithelial lesions into fibroadenomas and phyllodes tumors, which could be a valuable tool in assisting pathologists in differential diagnosis of breast fibroepithelial lesions.

Background

Fibroadenomas and phyllodes tumors are fibroepithelial lesions of the breast, characterized by proliferation of both epithelial and stromal components. Fibroadenomas are more commonly encountered on core biopsies than the rarer phyllodes tumors (approximately 20 % and <1 % of breast core needle biopsies respectively) [1, 2]. The preoperative distinction between the two lesions has significant impact on subsequent treatment. The current recommended management for phyllodes tumor diagnosed on core biopsy is wide excision without axillary staging regardless of grade [3]. Conversely, fibroadenomas are observed conservatively or, if tumors are larger than 2 cm, may be simply excised without achieving negative surgical margins [3]. This approach is due to the indolent behavior of fibroadenomas, despite sporadic reports of recurrences [4, 5], while phyllodes tumors have unpredictable outcomes with malignant tumors potentially progressing to metastasis and mortality [6–10]. It has been challenging separating cellular fibroadenoma from benign phyllodes tumor due to overlapping histological features, and this is particularly problematic on limited material of core biopsies, which may lead to over- or under-treatment for some patients, resulting in unnecessary anxiety and cost.

Several studies have proposed differentiating histological features such as stromal cellularity, stromal overgrowth, fragmentation, subepithelial condensation and presence of adipose tissue within stroma on core biopsies being indicative of phyllodes tumor [11–13].However, interpretation of these parameters is subjective, with interobserver variation and only moderate reproducibility between pathologists [11, 14]. Varied reports of immunohistochemical markers used in distinguishing phyllodes tumors from fibroadenomas suggest a lack of consensus and objectivity in assessing the expression of these biomarkers. Some authors reported Ki-67 expression to be helpful in diagnosing phyllodes tumors [15–17] but there are reports to the contrary [18, 19]. Lin et al. suggested a combination immunoscore of p16-INK4a and retinoblastoma-associated protein (pRB) [20] while Maity et al. reported expression of collagen I, III and CD105-positive microvessel density as parameters to differentiate the two lesions [21]. The vast majority of these studies were not conducted using pre-operative biopsies, which is where key management decision is required.

We set out to identify a useful molecular signature to help differentiate fibroadenomas from phyllodes tumors using pre-operative core biopsies to improve prediction of the final diagnosis.

Methods

Training set for assay development

The study received approval from the Centralized Institutional Review Board (CIRB 2005/002/F). As this was a retrospective study with anonymized cases, no specific patient consent was individually required. Forty-eight samples (24 fibroadenomas and 24 phyllodes tumors) were first employed as the training set for assay development. These included 10 paired core biopsies and surgical samples (20 samples), and 28 independent core and excisional samples from 38 patients (Table 1 and Additional file 1: Table S1). These formalin-fixed, paraffin-embedded (FFPE) samples were randomly selected from cases diagnosed at the Department of Pathology, Singapore General Hospital from 2008 to 2012. Hematoxylin and eosin (H&E)-stained slides were retrieved and reviewed. Phyllodes tumor was defined when there were well-developed fronds accompanied by increased stromal cellularity as opposed to fibroadenomas in which epithelial and stromal components were arranged in either intracanalicular or pericanalicular patterns without fronds or stromal hypercellularity. Differences in clinical features between fibroadenomas and phyllodes tumors were assessed with Mann–Whitney U test and Fisher’s exact test.

Table 1 Clinical features of the training cohort from 38 patients
Full size table

Expression profiling by Whole-Genome DASL® High Throughput (HT) Assay

Representative tumor areas were identified of which three to seven sections of 10-μm-thick sections from the same FFPE tumor block were obtained, deparaffinized and macrodissected. RNA was extracted using the RNeasy FFPE kit (Qiagen, Hilden, Germany) and quantified by Nanodrop Spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA). A total of 100 ng was used for quality assessment by real-time amplification of the RPL13A gene (forward primer, 5’-CACTTGGGGACAGCATGAG-3’, and reverse primer, 5’-GTAACCCCTTGGTTGTGCAT-3’) using the Power SYBR® Green RNA-to-CT™ 1-Step Kit (Life Technologies, Carlsbad, CA, USA) on a CFX96™ Real-Time PCR instrument (Bio-Rad Laboratories, Hercules, CA, USA). Samples with threshold cycle (Ct) below 29 were further subjected to quality assessment on a bioanalyzer. Eligible samples were submitted for expression profiling on the Whole-Genome DASL® HT Assay (Illumina, Inc., San Diego, CA, USA) at the Biopolis Shared Facilities A*Star, Singapore. The assay interrogates 29,377 features using the HumanHT-12 v4 BeadChip (Illumina, Inc.). Quantile-normalized gene expression data pre-analyzed using GenomeStudio® (Illumina, Inc.) was delivered. Data are available through GEO [GEO: GSE78071].

Selection of normalization genes and differentiating genes

Normalization genes were selected based on the smallest value of coefficient of variation among all samples. Differentiating genes were selected using the Significance Analysis of Microarrays package [22] and filtered based on the following criteria: (1) q-value less than 0.05; (2) mean difference of expression above 500; (3) R-fold above 1.5 (for genes highly expressed in phyllodes tumors) or less than 0.67 (for genes highly expressed in fibroadenomas).

Design of quantitative polymerase chain reaction (qPCR) assay

cDNA was synthesized from 1 μg of RNA using the High-Capacity cDNA Reverse Transcription Kit (Applied Biosystems®, Life Technologies, Carlsbad, CA, USA). Each qPCR assay consisted of 1X Power SYBR® Green PCR Master Mix (Life Technologies), 0.5 μM of forward and reverse primer each and 1 μl of 10-fold diluted cDNA as a template in a final total volume of 10 μl. Primers were designed using Primer-BLAST (NCBI, Bethesda, MD, USA) [23] with accession number listed in Additional file 1: Table S2. Non-template control acted as a negative control. Specificity of the amplicons was verified by melt curve analysis.

Data quantification and model building

Delta Ct (ΔCt) for each gene and sample was quantified as: ΔCt = Ct(gene, sample) − geomean(Ct(five normalization gene)). For comparison with expression of the Whole-Genome DASL® HT Assay, ΔCt data was transformed to 2-ΔCt as a positive linear scale and significance of correlation was analyzed with Pearson’s correlation test. Similarly to Cima et al. and Kälin et al. [24, 25], we used Random Forest (RF) ensemble classifier [26] to rank the importance of gene transcripts differentiating fibroadenomas and phyllodes tumors on qPCR assays. The top seven performing genes were used to build predictive logistic regression models using exhaustive search for the best model. To this end we used the glmulti package [27] with inclusion of the interaction terms. The best model was selected based on the lowest Akaike information criteria (AIC) value [28].

Validation cohort for model validation

The model of the multigene assay was tested on a separate set of 230 core biopsies with at least 2 years of follow-up. Hematoxylin and eosin (H&E)-stained slides were retrieved and reviewed. The outcome of the multigene assay was compared against the final diagnosis on the corresponding surgical excisions. Cases without subsequent surgical excisions were free from progression for at least 2 years and diagnosis made based on the initial core biopsy was used as the reference instead.

Results

Clinical features of training set

The clinical features and histology of the training set are shown in Table 1. Examples of the histological appearances of fibroadenoma and phyllodes tumor are shown in Fig. 1. Phyllodes tumors were significantly larger than fibroadenomas (p < 0.001). Median age of patients diagnosed with fibroadenomas and phyllodes tumors was 35 years and 44 years respectively (p = 0.09). No significant differences were observed for ethnicity distribution between the two groups of tumors.

Fig. 1
figure 1

Histology of fibroadenoma and phyllodes tumor on surgical excisions. a, b An example of a fibroadenoma at low-power (a) and high-power (b) magnifications. c, d An example of a benign phyllodes tumor at low-power (c) and high-power (d) magnifications

Full size image

Expression profiling and correlation with qPCR assays

Forty-seven samples (97.9 %) from 37 patients passed the quality control and were profiled successfully. Genes discriminating fibroadenomas and phyllodes tumors are listed in Additional file 1: Table S3. We designed and validated qPCR assays on 43 selected genes. Concordance between expression profiling and the qPCR assays was assessed based on a pilot run on six representative samples (Additional file 1: Table S4). Twenty-three assays with Pearson’s r of above 0.6 were further tested on the remaining 40 samples. One case was excluded due to insufficient material after expression profiling.

Development of a multigene qPCR panel

The results of ΔCt for all 23 qPCR assays were ranked using variable importance feature of the Random Forest classifier (Fig. 2). The seven most important genes in separating fibroadenomas and phyllodes tumors were TRIM29, FN1, CCL19, ABCA8, NPTX2, APOD and PRAME. A total of 268,435,456 candidate models were identified by glmulti based on these seven genes. We employed the genetic algorithm approach in the package to perform automated screening for the best model based on AIC value. A final five-gene model encompassing APOD, ABCA8, PRAME, FN1, and CCL19 with AIC of 14.2 was returned with coefficients as listed in Table 2.

Fig. 2
figure 2

Box plot of mean decrease in accuracy for 100 Random Forest trees generated for each gene. Mean decrease in accuracy measures the importance of each gene to the classification

Full size image
Table 2 Coefficients of the best model in predicting diagnosis in the 46-sample set
Full size table

Performance of the model

Patient demographics of the validation cohort of 230 core biopsies are shown in Table 3. Overall the assay was able to predict 213 (92.6 %) cases accurately. The prediction accuracy rates for fibroadenomas and phyllodes tumors were 94.7 % (179/189) and 82.9 % (34/41) respectively (Table 4), with positive (PPV) and negative (NPV) predictive values of 77.3 % and 96.2 %. A receiver operating characteristics curve with an area under the curve (AUC) of 0.948 (95 % confidence interval (CI) 0.913–0.983, p = 2.51E-19) was generated, indicating a large effect size in expected diagnostic performance for the five-gene assay (Fig. 3).

Table 3 Patient demographics of the validation cohort of 230 core biopsies
Full size table
Table 4 Performance of the five-gene model in predicting diagnosis in the independent validation cohort of 230 core biopsies
Full size table
Fig. 3
figure 3

Receiver operating characteristics (ROC) curve of the five-gene model with an area under the curve (AUC) of 0.948 (95 % confidence interval (CI) 0.913–0.983, p = 2.51E-19) in predicting diagnosis in the independent set of 230 core biopsies

Full size image

Of the 17 discordant cases (Table 5), seven were diagnosed as phyllodes tumors on pathological reports but were predicted as fibroadenomas on our assay. Upon review of these seven cases, two cases contained high epithelium content (Fig. 4), two were heterogeneous on histology with focal areas suggestive of fibroadenomas, while three other cases were confirmed as phyllodes tumors on review. The other ten of the 17 discordant cases were diagnosed as fibroadenomas on pathological reports but were predicted as phyllodes tumors on our assay. Among these ten cases, six cases had post-operative excisional material available as reference while the remaining four cases were benchmarked against the pre-operative pathological diagnosis. Of the six with excisional material, four were unequivocally fibroadenomas on histology, one was a cellular fibroadenoma without prominent fronds, and one was a fibroadenoma with sclerosing adenosis. Of the four pre-operative biopsies, one was unequivocally fibroadenoma, two cases contained features in keeping with fibroadenoma with hyalinized leafy fronds noted albeit without stromal cellularity, and one was an indeterminate case with focal areas of hemorrhage and high cellularity, which could not be definitively concluded on review.

Table 5 Seventeen cases with discordant outcomes between the five-gene assay and pathological diagnosis in the validation cohort. Post-operative diagnoses were used as benchmark reference unless otherwise stated
Full size table
Fig. 4
figure 4

Example of a discordant case containing high epithelium content. The five-gene assay predicted the core biopsies (a) as fibroadenoma but the final surgical excision (b) was diagnosed as phyllodes tumor on pathological reports

Full size image

Among the 230 core biopsies, the pre-operative pathological diagnoses were inconclusive for 22 cases where the term ‘fibroepithelial lesion’ was assigned, and there were three cases where the preoperative diagnoses were incongruous with the post-operative outcome (Table 6). Of these 25 cases, the five-gene assay was 80 % (20/25) accurate in classification with a PPV of 94.7 %.

Table 6 Cases with inconclusive pre-operative pathological diagnoses (n = 22), and discordant pre-and post-operative pathological diagnoses (n = 3). Among these cases, the five-gene assay was 80 % (20/25) accurate in predicting the final post-operative outcome
Full size table

Discussion

Classification of breast fibroepithelial tumors based on differentiating morphological and immunohistochemical features on pre-operative material has been challenging with variable findings across different groups (see Table 7 for summary). Jacobs et al. and Lee et al. first described individual pathological parameters which might help to differentiate fibroadenomas and phyllodes tumors in these limited samples [11, 16]. Jara-Lazaro proposed a combination of histological and immunohistochemical markers to indicate phyllodes tumors on core biopsies [15] but did not weigh the relative importance of each parameter in predicting phyllodes tumors. Morgan addressed this question by proposing a predictive tool including coefficient factors for each parameter to distinguish between fibroadenomas and phyllodes tumors but this has yet to be validated in an independent series of core biopsies [12]. Our study is the first to investigate differentiating features of fibroepithelial lesions on pre-operative material at the molecular level. We have developed a five-gene assay using a systematic approach based on genome-wide expression profiling data and validated the assay in an independent cohort of 230 pre-operative core biopsies of breast fibroepithelial lesions, the largest cohort reported so far. The pre-operative core biopsies were FFPE tissue containing low-quality RNA. Accordingly, our assay has been developed using RNA extracted from limited FFPE materials from core biopsies and thus is expected to perform on such material in the clinical setting.

Table 7 Summary of selected studies investigating differential features of fibroadenomas and phyllodes tumors on pre-operative materials
Full size table

Comparatively in surgical excisional materials, Huang et al. proposed a two-gene test derived from methylation profiling of an 11-gene panel in 86 samples [29], which described an elevated RASSF1A and/or TWIST1 methylation observed in phyllodes tumors as compared to fibroadenomas. They further evaluated the test in a separate validation cohort of 19 samples and reported a sensitivity and specificity of 0.33 and 0.75 respectively, with a PPV and NPV of 0.83 and 0.23. However, development of the test from a pre-selected panel of 11 genes may not be representative and the sample size of the validation cohort was too small to be conclusive. In contrast, our assay has a better sensitivity and specificity at 0.83 and 0.95 despite a lower PPV of 0.77. In a separate study by Kuijper interrogating the transcriptome differences between five fibroadenomas and eight phyllodes tumors, CTAG1/2, PRAME, HOXC13, ELF5 and FABP7 were among 96 other transcripts found to be highly differentially expressed between fibroadenomas and phyllodes tumors [30]. More recently, Vidal et al. reported a cluster of 47 epithelial- and luminal-related genes was found to be more expressed in fibroadenomas than phyllodes tumors among 105 breast cancer-related genes studied [31]. Findings from these studies however, were not further deployed as a test to distinguish fibroadenomas from phyllodes tumors on pre-operative materials despite the significant differential expression observed.

The training cohort comprised a mixture of surgical excisions and core biopsies with varying classifications of fibroepithelial lesions, simulating a realistic clinical scenario. Phyllodes tumors comprise benign, borderline and malignant grades on a continuous spectrum [32]. It is important that the assay works across the spectrum although one may argue that the malignant grade of phyllodes tumors is rarely in the histologic differential diagnosis between fibroadenomas and phyllodes tumors, and hence the assay may have little utility in the separation of fibroadenomas from malignant phyllodes tumors. The proportion of malignant phyllodes tumors included in the training cohort concurs with the incidence of malignant phyllodes tumors reported in the literature [33]. Nevertheless, even with the exclusion of malignant phyllodes tumors in the training cohort, differences of expression between fibroadenomas and phyllodes tumors for genes selected for the assay still fall within our selection criteria (R-fold differences above 1.5 and mean differences above 500) and hence would not have altered the assay development outcome. Excluding the three malignant tumor cases in the validation cohort only slightly reduces the sensitivity and PPV from 0.829 and 0.773 to 0.816 and 0.756 respectively. It is not our aim to investigate the differential expression between the phyllodes tumor grades although there is a trend of differences observed between grades in the expression of these five genes (results not shown). The sample sizes of borderline and malignant phyllodes tumors would be too small for meaningful analysis.

Several underlying factors which potentially limit the performance of the assay resulting in 17 discordant outcomes between the assay and pathological diagnosis include tumor heterogeneity and the issue of sampling on core biopsies. These factors may also have contributed to the three discordant pathological diagnoses between pre-operative core and post-operative excision materials. Core biopsies offer insight into only part of a tumor, which may not truly represent its entirety. Also, it is not uncommon for phyllodes tumors to contain areas indistinguishable from fibroadenomas, as seen in two discordant phyllodes tumor cases incorporating focal areas suggestive of fibroadenomas. Two other discordant phyllodes tumors harbored high epithelium content. The contribution of the epithelial component to the performance of the assay has yet to be ascertained although previous studies have shown that mutations were found in the stromal but not epithelial component [34, 35].

The limitation of our validation cohort is that the sample size for phyllodes tumor is small but the test was validated on a larger number of fibroadenomas, which have higher incidence compared to phyllodes tumors. We incorporated fibroadenomas on core biopsies which were not excised surgically although these may theoretically include uncertainty as the diagnoses are based solely on the core biopsy and not on the excised tumor. However, precluding fibroadenomas without subsequent excisions would result in a selection bias due to the exclusion of a large portion of representative cases. Moreover, the incidence of phyllodes tumor subsequent to a fibroadenoma diagnosis on core biopsy is very low [36], with an average duration of 12 months to the final correct diagnosis.

We do not advocate that the current diagnostic framework be replaced by the assay. Apart from the histological findings, clinical decision whether to proceed with surgical excision takes into account other factors such as radiological size and characteristics, as well as patient symptoms. For instance, a diagnosis of fibroadenoma on core biopsy may still be followed by excision if there is radiologic-pathologic discordance, or if the lesion is large or symptomatic. A diagnosis of phyllodes tumor on core biopsy however, warrants excision. Incorporating the results from our assay allows an additional tool that can be integrated into the decision-making process, enhancing precision especially when it affirms the pathological assessment on core biopsy. The gene assay is also helpful for pathologists in interpreting these lesions when the histological characteristics are indeterminate or ambiguous. This is exemplified by the 22 fibroepithelial lesions without a conclusive classification on core biopsy in the validation cohort. The multigene assay was able to classify 82 % of these cases accurately with a PPV of 94.7 %. The practicality and utility of the assay however, will need to be further validated in prospective studies.

The five-gene assay includes genes of various biological functions. FN1 (fibronectin 1) encodes a major component of the extracellular matrix. APOD (apolipoprotein D) and ABCA8 (ATP-binding cassette, sub-family A member 8) encode transporter proteins while PRAME (preferentially expressed antigen in melanoma) and CCL19 (chemokine ligand 19) genes are involved in immunoregulatory processes. Some of these genes were reported to be useful in differential diagnosis of other forms of tumors such as FN1 as a marker for renal cell carcinoma aggressiveness [37], PRAME as a marker for differentiating Müllerian carcinoma from malignant mesothelioma [38] and ABCA8 as part of a multigene gene assay for classifying cancer types [39]. While the individual functional role of these genes has not been implicated in breast fibroepithelial lesions, we found that these markers work best in combination for differential diagnosis between fibroadenomas and phyllodes tumors, as derived from our model algorithm. Nonetheless, it would be of interest to investigate the functional roles of these genes in breast fibroepithelial lesions in future studies.

Conclusions

We have developed a practical molecular assay for fibroepithelial lesions, classifying fibroadenomas and phyllodes tumors in pre-operative core biopsies. This may serve as an adjunctive aid for accurate pathological diagnosis. Prospective real-world trials will be helpful to determine whether improved surgical decision-making, supported by more accurate histological diagnosis, will lead to better outcomes.

Abbreviations

ABCA8 :

ATP-binding cassette, sub-family A member 8

AIC:

Akaike information criteria

APOD :

Apolipoprotein D

AUC:

area under the curve

CCL19 :

chemokine ligand 19

CI:

confidence interval

Ct:

threshold cycle

FA:

fibroadenoma

FEL :

fibroepithelial lesion

FFPE:

formalin-fixed, paraffin-embedded

FN1 :

fibronectin 1

H&E:

hematoxylin and eosin

HT:

high throughput

NPTX2 :

neuronal pentraxin II

NPV:

negative predictive value

PPV:

positive predictive value

PRAME :

preferentially expressed antigen in melanoma

pRB:

retinoblastoma-associated protein

PT:

phyllodes tumor

qPCR:

quantitative polymerase chain reaction

RF:

Random Forest

TRIM29 :

tripartite motif containing 29

References

  1. Dillon MF, Hill ADK, Quinn CM, O'Doherty A, McDermott EW, O'Higgins N. The accuracy of ultrasound, stereotactic, and clinical core biopsies in the diagnosis of breast cancer, with an analysis of false-negative cases. Ann Surg. 2005;242(5):701–7.

    Article  PubMed  PubMed Central  Google Scholar 

  2. Komenaka IK, El-Tamer M, Pile-Spellman E, Hibshoosh H. Core needle biopsy as a diagnostic tool to differentiate phyllodes tumor from fibroadenoma. Arch Surg. 2003;138(9):987–90.

    Article  PubMed  Google Scholar 

  3. National Comprehensive Cancer Network. Clinical practice guidelines in oncology. Breast cancer version 3.2014. http://www.nccn.org/professionals/physician_gls/pdf/breast.pdf. Accessed 2 September 2014.

  4. Grady I, Gorsuch H, Wilburn-Bailey S. Long-term outcome of benign fibroadenomas treated by ultrasound-guided percutaneous excision. Breast J. 2008;14(3):275–8.

    Article  PubMed  Google Scholar 

  5. Wang ZC, Buraimoh A, Iglehart JD, Richardson AL. Genome-wide analysis for loss of heterozygosity in primary and recurrent phyllodes tumor and fibroadenoma of breast using single nucleotide polymorphism arrays. Breast Cancer Res Treat. 2006;97(3):301–9.

    Article  CAS  PubMed  Google Scholar 

  6. Tan PH, Jayabaskar T, Chuah KL, Lee HY, Tan Y, Hilmy M, et al. Phyllodes tumors of the breast: the role of pathologic parameters. Am J Clin Pathol. 2005;123(4):529–40.

    Article  PubMed  Google Scholar 

  7. Tan PH, Thike AA, Tan WJ, Thu MMM, Busmanis I, Li H, et al. Predicting clinical behaviour of breast phyllodes tumours: a nomogram based on histological criteria and surgical margins. J Clin Pathol. 2012;65(1):69–76.

    Article  PubMed  Google Scholar 

  8. Parker SJ, Harries SA. Phyllodes tumours. Postgrad Med J. 2001;77(909):428–35.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  9. Belkacémi Y, Bousquet G, Marsiglia H, Ray-Coquard I, Magné N, Malard Y, et al. Phyllodes tumor of the breast. Int J Radiat Oncol Biol Phys. 2008;70(2):492–500.

    Article  PubMed  Google Scholar 

  10. Moffat CJ, Pinder SE, Dixon AR, Elston CW, Blamey RW, Ellis IO. Phyllodes tumours of the breast: a clinicopathological review of thirty-two cases. Histopathology. 1995;27(3):205–18.

    Article  CAS  PubMed  Google Scholar 

  11. Lee AHS, Hodi Z, Ellis IO, Elston CW. Histological features useful in the distinction of phyllodes tumour and fibroadenoma on needle core biopsy of the breast. Histopathology. 2007;51(3):336–44.

    Article  CAS  PubMed  Google Scholar 

  12. Morgan JM, Douglas-Jones AG, Gupta SK. Analysis of histological features in needle core biopsy of breast useful in preoperative distinction between fibroadenoma and phyllodes tumour. Histopathology. 2010;56(4):489–500.

    Article  PubMed  Google Scholar 

  13. Yasir S, Gamez R, Jenkins S, Visscher DW, Nassar A. Significant histologic features differentiating cellular fibroadenoma from phyllodes tumor on core needle biopsy specimens. Am J Clin Pathol. 2014;142(3):362–9.

    Article  PubMed  PubMed Central  Google Scholar 

  14. Lawton TJ, Acs G, Argani P, Farshid G, Gilcrease M, Goldstein N, et al. Interobserver variability by pathologists in the distinction between cellular fibroadenomas and phyllodes tumors. Int J Surg Pathol. 2014;22(8):695–8.

    Article  PubMed  PubMed Central  Google Scholar 

  15. Jara-Lazaro AR, Akhilesh M, Thike AA, Lui PC-W, Tse GM-K, Tan PH. Predictors of phyllodes tumours on core biopsy specimens of fibroepithelial neoplasms. Histopathology. 2010;57(2):220–32.

    Article  PubMed  Google Scholar 

  16. Jacobs TW, Chen Y-Y, Guinee JDG, Holden JA, Cha I, Bauermeister DE, et al. Fibroepithelial lesions with cellular stroma on breast core needle biopsy: are there predictors of outcome on surgical excision? Am J Clin Pathol. 2005;124(3):342–54.

    Article  PubMed  Google Scholar 

  17. Yohe S, Yeh IT. “Missed” diagnoses of phyllodes tumor on breast biopsy: pathologic clues to its recognition. Int J Surg Pathol. 2008;16(2):137–42.

    Article  PubMed  Google Scholar 

  18. Kaya R, Pestereli HE, Erdogan G, Gülkesen KH, Karaveli S. Proliferating activity in differential diagnosis of benign phyllodes tumor and cellular fibroadenomas: is it helpful? Pathol Oncol Res. 2001;7(3):213–6.

    Article  CAS  PubMed  Google Scholar 

  19. Barwijuk-Machala M, Musiatowicz B, Cylwik J, Reszec J, Augustynowicz A. AgNOR, Ki-67 and PCNA expression in fibroepithelial tumours of the breast in correlation with morphological features. Folia Morphol (Warsz). 2004;63(1):133–5.

    Google Scholar 

  20. Lin C-K, Tsai W-C, Lin Y-C, Yu J-C. Biomarkers distinguishing mammary fibroepithelial neoplasms: a tissue microarray study. Appl Immunohistochem Mol Morphol. 2014;22(6):433–41.

    Article  CAS  PubMed  Google Scholar 

  21. Maity PP, Chatterjee S, Das RK, Mukhopadhyay S, Maity A, Maulik D, et al. Finding an optimum immuno-histochemical feature set to distinguish benign phyllodes from fibroadenoma. Micron. 2013;48:34–41.

    Article  CAS  PubMed  Google Scholar 

  22. samr: SAM: significance analysis of microarrays. R package version 2.0. http://CRAN.R-project.org/package=samr. Accessed 13 October 2014.

  23. Ye J, Coulouris G, Zaretskaya I, Cutcutache I, Rozen S, Madden TL. Primer-BLAST: a tool to design target-specific primers for polymerase chain reaction. BMC Bioinforma. 2012;13:134.

    Article  CAS  Google Scholar 

  24. Kalin M, Cima I, Schiess R, Fankhauser N, Powles T, Wild P, et al. Novel prognostic markers in the serum of patients with castration-resistant prostate cancer derived from quantitative analysis of the pten conditional knockout mouse proteome. Eur Urol. 2011;60(6):1235–43.

    Article  PubMed  Google Scholar 

  25. Cima I, Schiess R, Wild P, Kaelin M, Schuffler P, Lange V, et al. Cancer genetics-guided discovery of serum biomarker signatures for diagnosis and prognosis of prostate cancer. Proc Natl Acad Sci U S A. 2011;108(8):3342–7.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  26. Liaw A, Wiener M. Classification and Regression by randomForest. R News. 2002;2(3):18–22.

    Google Scholar 

  27. glmulti: Model selection and multimodel inference made easy. R package version 1.0.7. http://CRAN.R-project.org/package=glmulti. Accessed 13 October 2014.

  28. Calcagno V, de Mazancourt C. glmulti: an R package for easy automated model selection with (generalized) linear models. J Stat Softw. 2010;34(12):1–29.

    Article  Google Scholar 

  29. Huang KT, Dobrovic A, Yan M, Karim RZ, Lee CS, Lakhani SR, et al. DNA methylation profiling of phyllodes and fibroadenoma tumours of the breast. Breast Cancer Res Treat. 2010;124(2):555–65.

    Article  CAS  PubMed  Google Scholar 

  30. Kuijper A. Pathogenesis and progression of fibroepithelial breast tumors. The Netherlands: Utrecht University; 2006.

    Google Scholar 

  31. Vidal M, Peg V, Galvan P, Tres A, Cortes J, Ramon y Cajal S, et al. Gene expression-based classifications of fibroadenomas and phyllodes tumours of the breast. Mol Oncol. 2015;9(6):1081–90.

    Article  CAS  PubMed  Google Scholar 

  32. Tan BY, Acs G, Apple SK, Badve S, Bleiweiss IJ, Brogi E, et al. Phyllodes tumours of the breast: a consensus review. Histopathology. 2016;68(1):5–21.

    Article  PubMed  PubMed Central  Google Scholar 

  33. Spitaleri G, Toesca A, Botteri E, Bottiglieri L, Rotmensz N, Boselli S, et al. Breast phyllodes tumor: a review of literature and a single center retrospective series analysis. Crit Rev Oncol Hematol. 2013;88(2):427–36.

    Article  PubMed  Google Scholar 

  34. Lim WK, Ong CK, Tan J, Thike AA, Ng CC, Rajasegaran V, et al. Exome sequencing identifies highly recurrent MED12 somatic mutations in breast fibroadenoma. Nat Genet. 2014;46(8):877–80.

    Article  CAS  PubMed  Google Scholar 

  35. Tan J, Ong CK, Lim WK, Ng CC, Thike AA, Ng LM, et al. Genomic landscapes of breast fibroepithelial tumors. Nat Genet. 2015;47(11):1341–5.

    Article  CAS  PubMed  Google Scholar 

  36. Jacobs TW, Chen YY, Guinee DG, Eby PR, Thike AA, Vohra P, Tan PH. Phyllodes tumor subsequent to a diagnosis of fibroadenoma on breast core needle biopsy: frequency and characteristics. In: USCAP 103rd annual meeting 2014: March 1–7 2014. San Diego, CA, USA: Modern Pathology; 2014. p. 56A.

    Google Scholar 

  37. Waalkes S, Atschekzei F, Kramer MW, Hennenlotter J, Vetter G, Becker JU, et al. Fibronectin 1 mRNA expression correlates with advanced disease in renal cancer. BMC Cancer. 2010;10:503.

    Article  PubMed  PubMed Central  Google Scholar 

  38. Brenne K, Nymoen DA, Reich R, Davidson B. PRAME (preferentially expressed antigen of melanoma) is a novel marker for differentiating serous carcinoma from malignant mesothelioma. Am J Clin Pathol. 2012;137(2):240–7.

    Article  CAS  PubMed  Google Scholar 

  39. Xu K, Cui J, Olman V, Yang Q, Puett D, Xu Y. A comparative analysis of gene-expression data of multiple cancer types. PloS One. 2010;5(10), e13696.

    Article  PubMed  PubMed Central  Google Scholar 

Download references

Acknowledgements

This study was supported by the Institute of Bioengineering and Nanotechnology (Biomedical Research Council, Agency for Science, Technology and Research, Singapore) and the Biomedical Research Council-Science and Engineering Research Council (BMRC-SERC) Joint Diagnostics Grant. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.We would like to thank Dr Vivien Koh of Biopolis Shared Facilities, Agency of Science, Technology and Research (A*Star), Singapore, for her contribution to microarray work.

Author information

Authors and Affiliations

  1. Division of Biodevices and Diagnostics, Institute of Bioengineering and Nanotechnology, 31 Biopolis Way, The Nanos, #04-01, Singapore, 138669, Republic of Singapore

    Wai Jin Tan, Igor Cima, Yukti Choudhury, Xiaona Wei & Min-Han Tan

  2. Department of Pathology, Singapore General Hospital, 20 College Road, Academia, Level 7, Diagnostics Tower, Singapore, 169856, Republic of Singapore

    Jeffrey Chun Tatt Lim, Aye Aye Thike & Puay Hoon Tan

  3. Duke-NUS Graduate Medical School, 8 College Road, Singapore, 169857, Republic of Singapore

    Puay Hoon Tan

Authors
  1. Wai Jin Tan
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Igor Cima
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Yukti Choudhury
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Xiaona Wei
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Jeffrey Chun Tatt Lim
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Aye Aye Thike
    View author publications

    You can also search for this author in PubMed Google Scholar

  7. Min-Han Tan
    View author publications

    You can also search for this author in PubMed Google Scholar

  8. Puay Hoon Tan
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding authors

Correspondence to Min-Han Tan or Puay Hoon Tan.

Additional information

Competing interests

MHT, PHT, WJT, and IC have filed for patents in molecular diagnostics with this assay. All other authors have no other competing interests.

Authors’ contributions

WJT performed RNA extraction, developed and validated the qPCR assays, analyzed the data, and drafted the manuscript. IC participated in model development, performed data analysis, and drafted the manuscript. YC designed the workflow of the study and participated in the revision of the manuscript. XW participated in the data analysis of WG-DASL and revision of the manuscript. JLCT collected FFPE samples, collated clinical data and helped to draft the manuscript. AAT participated in selection of cases and revision of the manuscript. MHT conceived and designed the study, coordinated all experiments, and critically revised the manuscript. PHT conceived and supervised the study, participated in case selection, and critically revised the manuscript. All authors read and approved the final manuscript.

Additional file

Additional file 1:

Supplementary tables. Table S1 Details of 48 samples from 38 patients constituting the training cohort. Table S2 Primers designed for potential differentiating genes and normalization genes. Table S3 Significant genes differentially expressed between fibroadenomas (FAs) and phyllodes tumors (PTs). Table S4 Correlation between expression profiling and qPCR assays based on a pilot run of six representative samples. Genes with good correlation value (r ≥ 0.6) were subjected to testing on remaining 40 samples. (XLSX 59 kb)

Rights and permissions

Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Tan, W.J., Cima, I., Choudhury, Y. et al. A five-gene reverse transcription-PCR assay for pre-operative classification of breast fibroepithelial lesions. Breast Cancer Res 18, 31 (2016). https://doi.org/10.1186/s13058-016-0692-6

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1186/s13058-016-0692-6

Keywords

Breast Cancer Research

ISSN: 1465-542X

Contact us