Several days ago, I had a new patient referred to me who has something I've never seen before. Though patients will occasionally ask about the risk of their lung cancer being spread by a biopsy or surgery, I've not seen and have heard of this only very, very rarely with lung cancer -- though it's not uncommon to see growth through the needle track of a biopsied mesothelioma.

The man in question had undergone resection of a left upper lobe tumor in early 2007, for which he underwent a lobectomy after a CT-guided biopsy demonstrated a moderately differentiated adenocarcinoma. His thorough mediastinoscopy followed by resection demonstrated no nodal involvement, and his scans were fine for two years. Nevertheless, he developed left shoulder pain a few months ago and eventually underwent a CT that demonstrated a soft tissue mass toward the top of his left chest, as well as enlarged to 1.5-2 cm behind his left collarbone (which would drain from the area where the chest was mass was found). He saw his surgeon, who thought it unlikely to be metastatic spread, or locoregional recurrence. Nevertheless, a biopsy of one of the supraclavicular nodes revealed an adenocarcinoma remarkably similar to his original cancer and with proteins present that indicate that they started as lung cells.

Looking at his imaging from today compared to his biopsy from 2007, it seems extremely likely that this local recurrence represents a new tumor in the needle track, along with nearby nodal drainage.

(click on image to enlarge)

His PET and head MRI show no evidence of disease outside of the left chest wall and lymph nodes near his left collarbone, so we are treating him with curative intent: specifically, concurrent chemo and radiation, and I'm very hopeful that he'll do well. But it's frustrating to have him face this challenge after probably being cured if not for needle track seeding.

At our thoracic oncology tumor board where we discussed his case, we talked about this being an approximately 1 in 1000 event, and obviously we think of this more with some other cancers. Still, I wonder if we'd see this more in NSCLC if more patients were able to be followed for years instead of the much lower ceiling we have for so many patients.

The last thing I want to do is suggest that this is something that people should worry about as a likely event after a biopsy. A CT-guided biopsy is often critical in diagnosing a cancer or sometimes proving that a suspicious lesion is not a cancer. I may never see another case in my career of someone with a situation so consistent with needle track seeding, but I've already had at least 8-10 cases of patients over the past decade who I thought were highly likely to have lung cancer but proved to have an infection or inflammation -- these were patients especially well served by getting a biopsy before we pursued a surgery, radiation, or chemotherapy.

Still, along with the more routine risks of a pneumothorax (collapsed lung), bleeding issues, or a very small risk of infection, this case illustrates that the risk of seeding from the biopsy procedure is not zero. It should also remind us why in medicine we so prefer to avoid using words like never. Still, medicine should be guided by a comparison of the balance of benefits and risks of competing alternatives, and it would be a mistake to weigh a remote risk too heavily when making decisions. A biopsy remains the way that we diagnose a cancer, and the risk of not knowing what a lung mass represents far exceeds a small risk of a serious complication from the biopsy itself.