Tumour cell displacement after 14G breast biopsy

L.E. Hoorntjea,*, M.E.I. Schipperb, A. Kayaa, H.M. Verkooijena,J.G. Klinkenbijlc, I.H.M. Borel Rinkesa

aDepartment of Surgery, University Medical Centre, P.O. Box 85500, 3508 GA Utrecht, The NetherlandsbDepartment of Pathology, University Medical Centre, P.O. Box 85500, 3508 GA Utrecht, The NetherlandscDepartment of Surgery, Rijnstate Hospital, Arnhem, The Netherlands

Accepted for publication 4 March 2004

KEYWORDSBreast cancer; Large core

needle biopsy; Needle

track; Non-palpable

Summary Introduction. Seeding of biopsy needle tracks with viable malignant cellswas an initial concern with all diagnostic breast needle procedures, including 14Gautomated needle biopsy. In an attempt to further evaluate this phenomenon, we haveaddressed the following questions: (1) are the tracks left by the needle biopsyprocedure detectable in the surgical excision specimen? (2) Are displaced tumour cellsvisible along the needle tracks? (3) Is it possible to identify and excise the entire needletrack for thorough histopathologic evaluation?

Methods. Surgical biopsy specimens of patients previously diagnosed with cancer onstereotactic 14G-needle biopsy were studied to find needle tracks. These arecharacterised by hemosiderin, scar tissue, foreign body giant cell reaction, fatnecrosis, or organizing haemorrhage. Occurrence of displaced tumour cells or groupsof cells along the tracks was registered.

Results. Needle tracks were found in 22/64 excision specimens of patients whounderwent 14G-needle biopsy and surgery on the same day. Tumour-cell displacementalong the needle track was seen in 11/22 cases (50%). In a prospective study, anattempt was made to excise the entire needle track in 13 consecutive cancer casesafter 14G biopsy. Median time interval between core biopsy and first surgical excisionwas 21 days (range 7–35). Needle tracks could be visualised in 11/13 cases; displacedcells were seen in seven.

Conclusion. Needle tracks can be found in the excision specimens of patients whopreviously underwent 14G biopsy, and displaced tumour cells can be recognised.Excision of the entire needle track is neither feasible, nor advisable as a routinemeasure, since radiotherapy is advised for both in situ and invasive carcinoma afterconservative surgery.Q 2004 Elsevier Ltd. All rights reserved.

Introduction

Image guided 14G automated needle biopsy isreplacing wire-localised excision for the evaluationof non-palpable lesions of the breast.

A disadvantage of this technique is that seedingof malignant tumour cells along the needle tracks

EJSO (2004) 30, 520–525

www.ejso.com

0748-7983/$ - see front matter Q 2004 Elsevier Ltd. All rights reserved.doi:10.1016/j.ejso.2004.03.001

*Corresponding author. Tel.: þ31-30-250-8074; fax: þ31-30-254-1944.

E-mail address: l.hoorntje@tiscali.nl

may occur recurrent cancer at the needle tracksites has been reputed after biopsy of prostate, lungor pancreatic cancer,1 –4 as for breast cancer.5 –9

This includes fine-needle aspirations, needle loca-lised excision biopsies, and large gauge needle corebiopsy.5,10 –15 The clinical significance of displacedtumour cells is unclear. Most publications reporteither the existence of displaced cells, or theoverall recurrence rate after diagnostic needleprocedures. No studies exist that include a detailedprospective registration of the position of theneedle track at the moment of 14G stereotacticbreast biopsy, nor have investigators attempted toexcise the entire needle track to study it in its fulllength.

This study was undertaken to evaluate needletrack seeding of tumour cells following large coreneedle biopsy. The following questions wereaddressed: (1) are the tracks left by the needlebiopsy procedure detectable in the surgical excisionspecimen? (2) Are displaced tumour cells visiblealong the needle tracks? (3) Is it possible to identifyand excise the entire needle track for thoroughhistopathologic evaluation?

Methods

Biopsy procedure

We use a prone table (Fisher Imaging Denver, CO),and a 14G-core needle, long throw (2.2 cm excur-sion) automated biopsy device with multiple passes(C.R. Bard Inc., Covington, GA). Lesions arelocalised with digital mammography. Our protocoladvises to take at least eight core biopsies in casesof calcifications and five in cases of density orarchitectural distortion. Our diagnostic protocolhas been described in detail elsewhere.16

Step 1: pilot

To assess the feasibility of finding and evaluatingthe needle tracks left by a stereotactic breastbiopsy procedure, we first carried out a pilot. Thehistological slides of 64 consecutive patients whounderwent surgical excision for a breast malignancypreviously diagnosed at 14G-needle biopsy wereevaluated. All patients underwent needle biopsyand surgical excision on the same day (this was theinitial protocol when stereotactic large core needlebiopsy was introduced in our clinic, as part of thelearning curve). Needle tracks were histologicallyidentified by the presence of recent bleeding, fatnecrosis, cellular debris, scar tissue, foreign body

giant cell reaction, or organising hemorrhage(Fig. 1(A)). In the cases where the needle trackcould be visualised, occurrence of tumour celldisplacement (single cells or groups of cells) alongthe needle track was studied (Fig. 1(B)). A tumourcell was considered to be displaced if not sur-rounded by basement membrane (stained withalpha-smooth muscle actin; Fig. 1(C)) or specialisedmammary stroma. In addition, tumour cells had tobe morphologically identical to the cells of theprimary carcinoma (stained with cam 5.2, anepithelial cell marker; Fig. 1(D)). Localisation ofthe displaced cells was registered as either insidethe tumour (Fig. 2) or outside (Fig. 3). Thisdistinction was made because displaced tumourcells within the tumour may have no clinicalconsequences. Tumour cells that are displacedoutside the tumour, may not be included in theexcision biopsy.

Step 2: prospective study

We wished to examine the entire needle track forhistopathological evaluation, through a pilot study.All patients scheduled to undergo stereotactic 14G-needle biopsy of a non-palpable breast lesion wereeligible for inclusion. During the biopsy-procedures,the course of the needle track was carefullymonitored and described. After written informedconsent, patients with invasive or in situ carcinomaon the large core specimen were included. The pre-operative localisation wire was positioned as closeas possible to and running in the same direction asthe needle track. The skin puncture site of theneedle track was included in the excision. Thespecimen was oriented on a wax plate and fixed informaline for 24 h. The resection margins wereinked with Alcian Blue. Following a surgical andhistopathological protocol, the exact direction ofthe needle track in the excision specimen wasdetermined and in cross-sections, the whole needletrack from skin to tumour was totally embedded inparaffin blocks. The conventional serial sectionswere stained with hematoxylin and eosin, and wereexamined by a breast pathologist to identify theneedle track and the occurrence, localisation andamount of tumour cell displacement, as describedearlier.

The amount of tumour cell displacement wasdivided in two groups: ‘single cell’ displacement, inwhich dislodgement of one to 10 tumour cells wasfound, and ‘cell clusters’, which contained dislod-gement of or ore more cell clusters, consisting ofmore than 10 cells per cluster. We assumed apositive relation between the amount of displaced

Tumour cell displacement after 14G breast biopsy 521

Figure 1 (A) Needle track surrounded by mammary and fatty tissue. (H & E, 10 £ ). (B) Detail with nests of vital tumourcells (arrows) growing along the needle track 18 days after stereotactic 14G core needle biopsy. Histological diagnosis ofthe tumour was a well differentiated invasive ductal carcinoma with lobular carcinoma in situ. (H & E, 100 £ ). (C) CAM-5.2 immunohistochemistry of the same area as shown in Fig. 1(B). (D) Alpha smooth muscle actin (a-sma) staining of thesame section as Fig. 1(B) demonstrating the absence of a basement membrane around the dislodged tumour cells(arrows).

Figure 2 (A) Displaced epithelial cells inside tumour (H & E, 2 £ ). (B) Detail (H & E, 100 £ ); arrows: displaced tumourcells.

L.E. Hoorntje et al.522

tumour cells and clinical consequences of malig-nant cell seeding.

Results

Pilot

In 22 of 64 excision specimens, the needle trackcould be identified. Tumour cell displacementalong the needle track was seen in 11 of the 22cases (50%) (Table 1). In four of these 11 cases, thedisplacement of tumour cells was also encountered

outside the tumour. Displacement of malignantcells had occurred in seven of 16 invasive tumoursand in four of six ductal carcinoma in situ (DCIS)lesions in which the needle track could be identified(Table 1).

Prospective study

In a series of 35 consecutive biopsy procedures,malignant lesions were present in 15 patients. Onepatient was lost to follow-up because she under-went surgery in another hospital. Another patientwas excluded because the needle track was not

Table 1 Retrospective assessment of excision specimens in which 14G needle tracks could be identified

Histology of surgical excision specimen Cases ðnÞ Occurrence of displaced tumour cells Displaced tumour cells outside tumour

Invasive 10 2 1Invasive þ DCIS 6 5 1DCIS 6 4 2Total 22 11 (50%) 4 (18%)

DCIS, ductal carcinoma in situ.

Figure 3 (A) Displaced groups of epithelial cells outside tumour (H & E, 2 £ ); white arrows: inked margin of resectionspecimen. (B) Detail (H & E, 100 £ ). (C) CAM5.2 staining of the same section as Fig. 3(B). (D) Alpha-sma of the samesection as Fig. 3(B).

Tumour cell displacement after 14G breast biopsy 523

excised for cosmetic reasons. Median time intervalbetween core biopsy and first surgical excision was21 days (range 7–35 days). The needle tracks left bycore biopsy could be visualised in 11 of these 13remaining surgical specimens (Table 2). Due totechnical problems, immunohistochemical stainingfailed in one case. Displaced tumour cells were seenin seven cases; these were located outside theprimary tumour in six patients. Displacement ofsingle cells was seen as frequent as displacement ofviable cell clusters. Tumour cells were displacedapproximately as often inside the tumour as outsideof it.

Discussion

This study shows that needle tracks after 14Gneedle biopsy can be recognised in the subsequentexcision specimen and that tumour cell displace-ment along these needle tracks does occur. Despitea special (radiological, surgical and histopathologi-cal) protocol the needle track could not be found intwo cases. Remains of needle tracks were found inre-excision specimens in two cases. Our resultssuggest that all displaced tumour cells cannot bedetected. The true incidence of tumour celldisplacement after 14G automated core biopsy isvery difficult to assess.

The clinical implications of finding malignanttumour cells along the needle track are unclear.Diaz et al. found that the incidence and amount oftumour cell displacement was inversely related tothe interval between core biopsy and excision.7

Their conclusion was that this relation suggests thattumour cells do not survive displacement. How-ever, Stolier et al. studied local breast cancerrecurrence in relation to mammographically guidedpunctures and found that one of two patients with

malignant needle track seeding developed a localrecurrence at 34 months,6 suggesting that displacedtumour cells may be viable.

Excising the needle track within the breastduring surgical therapy might prevent local recur-rence due to malignant cell seeding. In mastect-omy, this is rarely a problem, although seedingmight still occur occasionally in the overlying skinflaps. Concern about needle track seeding is greaterfor patients treated with breast-conservingtherapy. Since the approach in stereotactic corebiopsy may differ from the surgical routing, theneedle track often lies outside the planned surgicalexcision area. In addition, excising the needle trackmay not always be cosmetically preferable orpossible, as we found in our prospective study. Wedo not thus recommend excising the entire needletrack as a routine.

The use of radiotherapy may be adequate to killdisplaced viable cancer cells. Boutin et al. studiedpatients with mesothelioma who underwent inva-sive diagnostic procedures.17 Eight of 20 patientswho did not receive radiation therapy developedmetastases along the needle track vs. none of20 patients who did receive radiation therapy.Thurfjell et al. found displaced tumour cells afterfine needle aspiration and wire localised excisionin three of 33 patients with local recurrence ofbreast cancer.18 All three patients had notreceived radiotherapy. At present, the EORTCprotocol advises radiotherapy after breastconserving therapy for both invasive and in situcarcinoma.19

In conclusion, malignant tumour cells can bedisplaced during 14G-needle biopsy of the breast.Excision of the entire needle track is neitherfeasible, nor advisable as a routine measure sinceradiotherapy is advised for both in situ and invasivecarcinoma after consevative surgery.

Table 2 Prospective study attempting to excise the entire needle track for evaluation

Histology Cases ðNÞ Displaced tumour cells insidetumour

Displaced tumour cells outsidetumour

Intervala (days)

Single cells Cellclustersb Single cells Cellclustersb

IC 4 2 2 3 2 18–35IC þ DCIS 7c 1 2 3 3 7–25DCIS 2d 0 0 0 0 21–23Total 13 3 (23%) 4 (31%) 6 (46%) 5 (38%)

IC, invasive carcinoma; DCIS, ductal carcinoma in situ.a Interval between corebiopsy and excision.b Cell clusters defined as .10 cells.c In 1 case, no needle track could be identified in the excision specimen.d In 1 case staining failed due to technical error.

L.E. Hoorntje et al.524

Acknowledgements

LEH’s position was financed by a grant from theDutch Scientific Research Committee (NWO)—Medi-cal Sciences stipend no 920-03-159 (AGIKO).

References

1. Wolinsky H, Lischner MW. Needle track implantation oftumor after percutaneous lung biopsy. Ann Intern Med 1969;71(2):359—362.

2. Ferrucci JT, Wittenberg J, Margolies MN, Carey RW.Malignant seeding of the tract after thin-needle aspirationbiopsy. Radiology 1979;130(2):345—346.

3. Moul JW, Miles BJ, Skoog SJ, McLeod DG. Risk factors forperineal seeding of prostate cancer after needle biopsy.J Urol 1989;142(1):86—88.

4. Ishii H, Okada S, Okusaka T et al. Needle tract implantationof hepatocellular carcinoma after percutaneous ethanolinjection. Cancer 1998;82(9):1638—1642.

5. Harter LP, Curtis JS, Ponto G, Craig PH. Malignant seeding ofthe needle track during stereotaxic core needle breastbiopsy. Radiology 1992;185(3):713—714.

6. Stolier A, Skinner J, Levine EA. A prospective study ofseeding of the skin after core biopsy of the breast. Am J Surg2000;180(2):104—107.

7. Diaz LK, Wiley EL, Venta LA. Are malignant cells displaced bylarge-gauge needle core biopsy of the breast? AJR Am JRoentgenol 1999;173(5):1303—1313.

8. Bott S, Mohsen V, Wells C, Carpenter R. Needle tract meta-stases in breast cancer. Eur J Surg Oncol 1999;25(5):553.

9. Roussel F. Risk of metastasis during fine needle aspiration.J Clin Pathol 1990;43(10):878—879.

10. Roussel F, Dalion J, Benozio M. The risk of tumoral seeding inneedle biopsies. Acta Cytol 1989;33(6):936—939.

11. Grabau DA, Andersen JA, Graversen HP, Dyreborg U. Needlebiopsy of breast cancer. Appearance of tumour cells alongthe needle track. Eur J Surg Oncol 1993;19(2):192—194.

12. Youngson BJ, Cranor M, Rosen PP. Epithelial displacement insurgical breast specimens following needling. Am J SurgPathol 1994;18(9):896—903.

13. Youngson BJ, Liberman L, Rosen PP. Displacement ofcarcinomatous epithelium in surgical breast specimensfollowing stereotaxic core biopsy. Am J Clin Pathol 1995;103(5):598—602.

14. Liberman L, Vuolo M, Dershaw DD et al. Epithelial displace-ment after stereotactic 11-gauge directional vacuum-assisted breast biopsy. AJR Am J Roentgenol 1999;172(3):677—681.

15. Fajardo LL. Breast tumor seeding along localization guidewire tracks. Radiology 1988;169(2):580—581.

16. Verkooijen HM. Diagnostic accuracy of stereotactic large-core needle biopsy for nonpalpable breast disease: results ofa multicenter prospective study with 95% surgical confir-mation. Int J Cancer 2002;99(6):853—859.

17. Boutin C, Rey F, Viallat JR. Prevention of malignant seedingafter invasive diagnostic procedures in patients with pleuralmesothelioma. A randomized trial of local radiotherapy.Chest 1995;108(3):754—758.

18. Thurfjell MG, Jansson T, Nordgren H, Bergh J, Lindgren A,Thurfjell E. Local breast cancer recurrence caused bymammographically guided punctures. Acta Radiol 2000;41(5):435—440.

19. Julien JP, Bijker N, Fentiman IS et al. Radiotherapy in breast-conserving treatment for ductal carcinoma in situ: firstresults of the EORTC randomised phase III trial 10853. EORTCBreast Cancer Cooperative Group and EORTC RadiotherapyGroup. Lancet 2000;355(9203):528—533.

Tumour cell displacement after 14G breast biopsy 525