Percutaneously implanted markers in peripheral lung tumours: Report of complications

To the Editor,

Lung tumours move during breathing, and tumour motion of more than 3 cm has been seen for tumours located near the diaphragmatic domes [1]. Breathing-adapted radiotherapy such as respiratory beam gating or tumour tracking, rely on the ability to determine and predict the breathing-related tumour motion based on an external or internal surrogate for tumour motion [2]. The prediction of the correlation between the tumour and the surrogate positions must be verified throughout the treatment; the verification can be performed with repeated kV-imaging of the tumour. However, not all lung tumours are well-defined on kV-images and therefore radio-opaque markers implanted in or close to the tumour have been used as a surrogate for tumour position. Markers can be implanted percutaneously guided by fluoroscopy or computed tomography (CT) or trans-bronchially inserted in nearby small bronchi. The advantage of percutaneous implantation is the possibility to implant the marker directly into the tumour assuring a good representation of tumour motion, but potentially at the cost of morbidity due to the risk of pneumothorax. We report the results of a small pilot study examining the feasibility of CT-guided percutaneous implantation of fiducial markers.

Material and methods

Fifteen patients were prospectively included between February 2009 and December 2010. Data collection was approved by the Danish Data Protection Agency (j.nr. 30-0484). The study was approved by the local ethical committee (j.nr. H-B-2007-016) and reported to Clinicaltrials.gov (ClinicalTrials.gov Identifier: NCT00910546).

Study inclusion criteria were: referral for SBRT, non-small cell lung cancer or lung metastases, age > 40 years, WHO performance status (PS) < 2 and obtained signed informed consent. Exclusion criteria were: forced expiratory volume in one second (FEV1) < 0.5 l, centrally located tumour, or tumour located close to large vessels.

All marker implantations were performed CT-guided by the same radiologist one week prior to treatment planning and two weeks prior to treatment start. Markers and implantation devices were changed during the study. For the first five patients a Visicoil^TM (IBA Dosimetry, Bartlett, TN, USA) gold marker measuring 0.75 mm × 2 cm was implanted with an 18 gauge needle (outer diameter 1.27 mm). Patient no. 6 had a complex helical platinum coil (Boston Scientific, Natick, MA, USA) measuring 20 mm restrained, 2 × 4× 4 mm unrestrained, implanted using an 18 gauge needle. Patients no. 7–15 had a Gold Anchor^TM marker (Naslund Medical, Huddinge, Sweden) measuring 0.28 × 20 mm implanted using a 25 gauge needle (outer diameter 0.53 mm). An improved inner coating of the preloaded needle was later introduced and used for patients no. 11–15. In all patients, only a single marker was implanted. All markers were complex in structure (Figure 1).

Figure 1. Photo showing the complex helical platinum marker (top), the Gold Anchor^TM marker (middle) and the Visicoil^TM gold marker (bottom).

After implantation a thoracic CT was acquired to check for pneumothorax. The patients were observed for four hours followed by a chest x-ray. In case of pneumothorax, a thoracic surgeon was consulted regarding further intervention and/or observation.

The planning protocol for stereotactic body radiation therapy (SBRT) consisted of a positron emission tomography (PET)/CT with contrast enhancement and, within two subsequent days, a four-dimensional (4D)CT and a voluntary deep inspiration breathhold CT (BHCT) in the same session. Treatment planning was performed on the midventilation phase of the 4DCT and a total dose of 45 Gy was delivered in three fractions within 5–8 days.

Results

Patient characteristics are presented in Table I. The marker of patient no. 2 disappeared between the planning scan and the treatment start. It had been placed in proximity to a bronchus and the patient had been reported coughing. The marker was probably expectorated, as a full-body CT revealed no sign of the marker.

Table I. Patient characteristics.

Age*		73 (60–87) years
Gender	Female/Male	7/8
FEV1*		1.4 (0.7–3.4) l
Diagnosis	NSCLC:
	Primary	11
	Recurrent	3
	Lung metastasis	1
Prior lung resection	Wedge resection	2
	Lobectomy	1
Tumour location	Upper lobe	6
	Lower lobe	9
Tumour size*	Volume	13.6 (1.7–46.3) cm^3
Tumour motion*	Left-Right	0.2 (0.1–0.3) cm
	Antero-Posterior	0.2 (0.1–0.4) cm
	Cranio-Caudal	0.6 (0.1–2.4) cm

*Median (range).

Patients no. 4, 5, 7 and 10 experienced a pneumothorax. None of the patients had symptoms of pneumothorax (lowering of blood oxygen saturation, increased breathing rate, dyspnoea or cough). Nevertheless three of the patients had a more than 3 cm collapse of the lung on x-ray and required a chest tube. For patients no. 7 and 10 the marker was initially not properly detached and withdrawn with the needle. A new marker was placed within the same session. The collapsed lung of patient no. 4 had not fully unfolded at the planning session and the patient had to be readmitted for a second chest tube thereby delaying treatment for a week. Hospitalisation time was 1–6 days (median 2.5 days). Patient no. 3 had haemoptysis before and after the marker implantation. In patient no. 6, the only patient with the complex helical platinum coil, the marker was implanted more than 1 cm below the tumour (Figure 2).

Figure 2. Chest x-rays showing different marker types. Patient 1 had a 2 cm long helical gold marker (Visicoil^TM) implanted in the tumour in the right inferior lung lobe (left). Patient 6 had the complex helical platinum marker (Boston Scientific) implanted in the tumour in right superior lung lobe (middle). Note that the marker was placed outside the tumour. The patient had a pacemaker. Patient 11 had the Gold Anchor^TM marker implanted in the tumour in the right inferior lobe (right). The arrows indicate the position of the markers.

By visual evaluation of the thoracic orthogonal kV x-rays acquired four hours after the marker implantation (Figure 2), it was clear that the Visicoil marker provided the highest radio-opacity. The two other markers provided equal radio-opacity. However, the radio-opacity of the Gold Anchor^TM marker depended on the intra-tumoral folding of the marker.

Discussion

In our study four of 15 patients (2/5 with Visicoil and 2/9 with Gold anchor) developed pneumothorax after CT-guided percutaneously marker implantation. In the two patients that developed pneumothorax after the 25 gauge needle implantation, the implantation was performed twice, since the marker was initially withdrawn with the needle, indicating that implantation of more markers may increase the risk of pneumothorax. A larger ratio of patients developed pneumothorax when implantation was done using the larger 18 gauge needle than with the thinner 25 gauge needle although not significantly in this small sample size.

The pneumothorax rate in this study did not exceed what previous studies using percutaneous implantation have reported [3–11], except for the study by de May et al. [8] where a pneumothorax rate of only 10% was reported. Indeed in a study by Kupelian et al. [12], also including 15 patients for percutaneous implantation of Visicoil^TM markers, eight of the patients had pneumothorax. The high pneumothorax rate requires that the benefit of marker implantation is substantial in terms of increased precision in treatment delivery and decreased toxicity to normal tissue with breathing-adapted radiotherapy. Planning studies have shown potential clinical benefit of breathing-adapted radiotherapy. However advanced breathing adaptation, e.g. respiratory gating or tumour tracking without markers and extensive image guidance, is difficult [13]. The real clinical benefit of breathing-adapted (marker guided) radiotherapy needs yet to be confirmed in large trials.

A limitation of the present study is the limited number of patients included and the long inclusion period. Although the radiologist implanting the markers was trained in interventional procedures and had performed hundreds of CT-guided biopsies, a learning curve can be expected when performing a new procedure, as he was not familiar with marker implantation prior to this present study. Thus the complication rate may be overestimated in this small study. The use of different markers represented an unscheduled modification of the clinical protocol. The gold Visicoil^TM marker gave the best radio-opacity in kV-imaging but the implantation needle was an 18 gauge needle and resulted in two cases of pneumothorax after just five implantations, why the use was omitted. The next marker, the complex helical platinum marker, was not preferred by our radiologist due to manual loading into a needle, in contrast to the two other types of markers that were preloaded. Last we used the Gold Anchor^TM marker which was inserted through a thin 25 gauge needle minimising the risk of pneumothorax. The thin needle was very flexible making the steering of the needle challenging. In our opinion the optimal marker for IGRT and insertion method need yet to be found.

Overall, we conclude that percutaneous implantation of markers was feasible but associated with a significant risk of pneumothorax. Larger clinical trials with continuous implantation of markers for breathing adaptation are warranted to estimate a reliable complication rate and to further improve breathing-adapted radiotherapy of lung cancer.

Acknowledgements

The authors have received grants from the Danish Council for Independent Research in Medical Sciences, The Arvid Nilssons Foundation, The Astrid Thaysens Foundation and the Lundbeck Foundation Center for Interventional Research in Radiation Oncology (CIRRO). The authors wish to thank Ingemar Näslund for providing the Gold Anchor^TM implantation devices.

Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.