When Lung Cancer Shrinks but Surgery Turns Dangerous: One Case’s Warning

A 64-year-old woman with lung adenocarcinoma received three cycles of nivolumab plus carboplatin and paclitaxel before her operation. Her tumor responded well by every oncologic measure that matters to medical oncologists: at surgery, pathologists found no viable cancer anywhere in the lung or the removed lymph nodes, a finding known as pathologic complete response. Yet the operation nearly killed her. While surgeons tried to peel a bulky lymph node away from her aortic arch, catastrophic bleeding erupted from the vessel before planned vascular control could be achieved, forcing an emergency conversion to cardiopulmonary bypass, deep hypothermic circulatory arrest, and removal and reconstruction of part of the aortic arch itself. The case, published as a case report in General Thoracic and Cardiovascular Surgery Cases, delivers an uncomfortable lesson: a tumor can disappear on the microscope and still leave behind tissue that makes surgery treacherous.

Contrast-enhanced computed tomography revealed a 16-millimeter solid tumor in the left upper lobe of her lung and, more ominously, a bulky lymph node in the station numbered 5, the subaortic station just beneath the aortic arch. That node had lost its fat plane with the arch, a radiographic sign suggesting the node was invading or was at least intimately adherent to the great vessel. Fluorodeoxyglucose positron emission tomography showed intense uptake in both the primary tumor and the node, with no evidence of disease elsewhere in the chest or body. An attempt to sample the node with transbronchial needle aspiration failed because the inflamed node was rigid, so surgeons instead performed a thoracoscopic wedge biopsy of the lung lesion itself.

Under the 9th edition of the Japanese General Rule for Clinical and Pathological Record of Lung Cancer, the disease was staged cT1bN2aM0, stage IIIA, with single-station N2a nodal involvement. That staging matters because N2 disease, meaning cancer in mediastinal lymph nodes, sits at the boundary of surgical resectability, and bulky or invasive N2 disease is generally classified as unresectable or reserved for exceptional multidisciplinary review in contemporary consensus statements.

The treatment team chose neoadjuvant nivolumab plus chemotherapy with curative resection as the intended strategy. She received carboplatin at an area under the curve of 5, paclitaxel at 200 milligrams per square meter of body surface, and nivolumab at 200 milligrams per body, for three cycles. No immune-related adverse event occurred. Restaging imaging showed no recurrence at the biopsied primary site and only slight shrinkage of the subaortic node, from 23 to 18 millimeters in short axis, a 14 percent reduction that qualified as stable disease under the Response Evaluation Criteria in Solid Tumors, the standard yardstick for measuring tumor response on scans.

Despite stable disease and persistent suspicion of aortic invasion, the multidisciplinary tumor board recommended proceeding with curative-intent resection. Their reasoning was specific: there had been no distant or nodal progression, the disease remained confined to a single N2a station, and the node was judged potentially completely resectable with cardiovascular backup available. The patient was informed of the elevated operative risk, including the possibility of great-vessel resection under cardiopulmonary bypass, and consented. She was positioned in a modified right lateral decubitus, or corkscrew, position that would allow simultaneous access to the left chest and the femoral vessels for emergency bypass cannulation. Surgery began through a left posterolateral thoracotomy, with aortic arch replacement planned if invasion by the node was confirmed.

It was during dissection over the aorta that the plan collapsed. The subaortic lymph node proved broadly and densely adherent to the aortic arch, the left recurrent laryngeal nerve, and surrounding mediastinal tissues. As surgeons attempted to separate it, catastrophic bleeding occurred from the aortic arch at the suspected invasion site, before the planned vascular control had been established. Manual compression failed. The team placed an arterial return cannula in the right femoral artery and a venous drainage cannula in the inferior vena cava, established cardiopulmonary bypass, cooled the patient’s core temperature to 21 degrees Celsius, and induced total circulatory arrest to expose the bleeding point. Only then could they see the damage: the injured arch had dehisced over approximately half of its circumference. Selective cerebral perfusion protected the brain while surgeons completed a left upper lobectomy with ND2a-1 lymph node dissection, en bloc removal of the node and aortic arch, and graft reconstruction. Operative time was 725 minutes and blood loss reached 8,389 milliliters.

The patient survived. A pleural complication on postoperative day 9, classified as Clavien-Dindo grade IIIa, resolved after thoracentesis, and she was discharged on day 20 without neurologic sequelae. Histopathology confirmed the oncologic success: no viable residual tumor in the lung or lymph nodes, 0 percent viable tumor, an Ef.3 grade of regression, and an R0, or complete, resection. But the aortic-wall specimen told the more instructive story. Lymphocyte-predominant inflammatory-cell infiltration extended to the aortic adventitia, the outermost layer of the vessel, and became continuous with the surrounding stroma. In other words, even with every cancer cell eliminated, the immune response and fibrosis triggered by treatment had welded the node to the aorta, leaving a hazardous mediastinal interface rather than the clean dissection plane surgeons had hoped a complete response would restore.

The authors argue that this case exposes a practical gap between oncologic response and technical resectability. In retrospect, they write, the limited nodal shrinkage and the persistent suspicion of aortic involvement were themselves indicators of a technically demanding operation. The key limitation was not that imaging failed to signal operative risk, but that current imaging modalities cannot reliably distinguish viable tumor from treatment-related fibrosis, nor accurately define the surgical dissection plane. Radiographic response, stable disease, or eventual pathologic complete response should therefore not be interpreted as evidence that dissection will be technically safe. This caution aligns with a recent multisocietal consensus holding that bulky and invasive N2 disease carries high operative risk and should generally be regarded as unresectable. When such disease is accompanied by suspected extension toward the aorta, the question is not only whether viable tumor remains but whether fibrosis, whether treatment-related or tumor-related, has fixed the node to critical structures.

The report also offers concrete guidance for teams facing similar dilemmas. Persistent loss of the fat plane or deformation of the aortic contour on restaging scans should be treated as warning signs of a potentially inseparable interface, regardless of RECIST response. Dedicated arterial-phase CT angiography can map arch-branch anatomy, vascular access, and potential clamping or endograft landing zones, and cine MRI may add information when CT findings are equivocal, though neither modality has been validated to distinguish viable tumor from post-chemoimmunotherapy fibrosis or to demonstrate a safe dissection plane. The authors recommend converting cardiovascular backup into a predefined control strategy that specifies exposure, proximal and distal control, cannulation readiness, and criteria for abandoning subadventitial peeling before the operation begins.

One emerging option, thoracic endovascular aortic repair, or TEVAR, may provide endoluminal protection and facilitate en bloc aortic-wall resection without cardiopulmonary bypass, but the evidence base is thin. This off-label oncologic use is supported only by small retrospective series, and selection requires suitable access, adequate disease-free landing zones, and planned preservation or revascularization of the supra-aortic branches, which makes TEVAR considerably less straightforward at the arch than in the descending aorta. Reported risks include endoleak, branch-vessel injury, stroke, spinal cord ischemia, graft migration or herniation, and infection, so the authors recommend considering it case by case within an experienced aortic-thoracic team rather than routinely. Their bottom line is blunt: if safe vascular control and complete resection cannot both be anticipated, definitive nonsurgical therapy remains appropriate. The broader context is a rapidly changing field. Neoadjuvant immunotherapy combined with platinum chemotherapy became an important option for resectable non-small cell lung cancer after the CheckMate 816 trial demonstrated higher pathologic complete response rates than chemotherapy alone, with similar benefit reported in Japanese patients. But those trials enrolled patients whose disease was already resectable, so they never answered whether bulky or invasive N2 disease can be safely resected after treatment. Historically, resection involving the thoracic aorta has been limited to carefully selected patients, and when complete resection is achieved in T4N0-1M0 disease, acceptable perioperative mortality and long-term survival have been reported. This single case, with the patient alive without recurrence 22 months after surgery, does not establish outcomes for others, but it documents vividly that a pathologic complete response and a safe operation are not the same thing, and that surgical planning should preserve caution whenever great-vessel invasion is suspected.

Subject of Research: A case report of surgery after neoadjuvant chemoimmunotherapy for bulky, invasive N2 lung cancer with suspected aortic arch invasion

Article Title: Lessons from surgery after neoadjuvant chemoimmunotherapy for bulky, invasive N2 lung cancer: a case report

Article References: Hashimoto, T., Tanaka, K., Fujita, Y., Manabe, T., Matsumiya, H., Mori, M., Takenaka, M., Kuroda, K., Tanaka, F., & Uramoto, H. (2026). Lessons from surgery after neoadjuvant chemoimmunotherapy for bulky, invasive N2 lung cancer: a case report. General Thoracic and Cardiovascular Surgery Cases, 5(1), Article 34. https://doi.org/10.1186/s44215-026-00271-4

Image Credits: AI Generated

DOI: 10.1186/s44215-026-00271-4

Keywords: lung cancer, NSCLC, neoadjuvant chemoimmunotherapy, nivolumab, N2 lymph nodes, pathologic complete response, aortic arch, cardiopulmonary bypass, surgical resectability, case report, RECIST, fibrosis