Systemic AL (λ) Amyloidosis Discovered After Neoadjuvant Pembrolizumab‐Based Chemoimmunotherapy for Resectable NSCLC: Case Report Authors: Weijie Ma, Dana M. Ferrari‐Light, Liang Lu Published: 2026 ABSTRACT ABSTRACTSystemic AL amyloidosis is rarely reported in temporal association with immune checkpoint inhibitor use. We report a 69‐year‐old man with resectable stage IIB right upper lobe lung adenocarcinoma who received neoadjuvant pembrolizumab, carboplatin, and pemetrexed followed by robotic‐assisted lobectomy. Pathology showed a 4.6‐cm treated tumour bed with residual invasive adenocarcinoma (50% viable), negative margins, and no nodal metastasis (0/9). Tumour profiling demonstrated a PD‐L1 tumour proportion score of 100%, a high tumour mutational burden (18 mut/Mb), microsatellite stability, and variants in BRAF, TP53 and PTPRT. In addition to the treatment effect, widespread Congo red–positive deposits were identified in lung parenchyma and multiple nodal stations. Laser microdissection with LC–MS/MS confirmed AL (λ) amyloid. Subsequent workup revealed a λ‐restricted plasma cell clone (6.4%) with t (11;14), establishing systemic AL amyloidosis. He received adjuvant pembrolizumab and daratumumab‐CyBorD with partial hematologic response. This case highlighted that amyloid can unexpectedly be a second diagnosis after post‐neoadjuvant lung resections and that proteomic subtyping is essential for prompt haematologic staging and treatment. Introduction Immune checkpoint inhibitors (ICIs) are increasingly incorporated into perioperative regimens for resectable non–small cell lung cancer (NSCLC). Rare immune‐related toxicities and unexpected comorbid diagnoses are therefore increasingly encountered along curative‐intent pathways [1]. Amyloidosis reported in temporal association with ICI therapy is uncommon and has most often involved the AA subtype, whereas systemic immunoglobulin light‐chain (AL) amyloidosis appears much less frequently reported in this setting [2]. We report a case in which systemic AL (λ) amyloidosis was first recognised incidentally in lung parenchyma and lymph nodes at lobectomy after neoadjuvant pembrolizumab‐based chemoimmunotherapy for stage IIB lung adenocarcinoma, prompting definitive proteomic typing and subsequent hematologic workup. Case Report A 69‐year‐old man with hypertension, gastroesophageal reflux disease, hypothyroidism, anxiety/depression, and a remote history of tonsillar cancer treated with tonsillectomy and chemoradiation (2005) presented in July 2024 with chest tightness and mild exertional dyspnoea. CT chest revealed a 6.4‐cm RUL mass. PET demonstrated FDG uptake confined to the RUL lesion without distant disease. CT‐guided biopsy on August 30, 2024, confirmed lung adenocarcinoma. Endobronchial ultrasound‐guided sampling of lymph node stations 4 L, 7, 4R and 10R was negative for malignancy. The pretreatment CT‐guided biopsy and EBUS‐guided nodal sampling were performed at an outside hospital. A review of the available outside pathology and cytology reports did not describe amyloid deposition in the pretreatment specimens. However, archival pretreatment tissue was not available for additional Congo red staining or proteomic analysis, and pretreatment serum was not available for retrospective light‐chain testing. Brain MRI was reportedly negative for metastatic disease. The tumour was staged cT3N0M0 (stage IIB). The patient received neoadjuvant pembrolizumab, carboplatin, and pemetrexed for four cycles and tolerated therapy well. Restaging PET/CT in December 2024 demonstrated a persistently FDG‐avid RUL lesion measuring approximately 5.0 × 3.2 cm, modestly decreased from baseline, with no nodal or distant metastatic disease (Figure 1). MRI of the right brachial plexus in December 2024 showed an apical RUL mass abutting the mediastinum without extension into the upper chest wall or neck. Clinical timeline and radiographic response to neoadjuvant chemoimmunotherapy. Schematic timeline of diagnosis, treatment, and subsequent evaluation for systemic AL amyloidosis in a patient with resectable right upper lobe (RUL) lung adenocarcinoma treated with neoadjuvant pembrolizumab plus carboplatin and pemetrexed (four cycles), followed by surgical resection and adjuvant pembrolizumab. Representative axial FDG PET/CT images demonstrate an FDG‐avid RUL mass at baseline (left) with an interval decrease in lesion size/avid focus after completion of neoadjuvant therapy (right), without evidence of distant disease on restaging. Robotic‐assisted RUL lobectomy with lymphadenectomy was performed in January 2025. The operation was technically challenging due to extensive adhesions and tumour adherence near the apex and subclavian vessels. Histopathologic evaluation demonstrated residual invasive lung adenocarcinoma in a 4.6‐cm treated tumour bed with marked treatment effect (Figure 2A). Architectural patterns included complex glandular (cribriform/fused) predominantly (40%), with mixed acinar (30%) and solid (30%) components (Figure 2B,C). Visceral pleural invasion was present. Spread through air spaces and lymphovascular invasion were not identified. Treatment effect quantification showed 50% residual viable tumour, 2% necrosis, and 48% stroma (fibrosis and inflammation). Nine lymph nodes were negative for carcinoma (0/9). Tumour biomarker profiling showed a PD‐L1 tumour proportion score of 100%, a high tumour mutational burden (18 mut/Mb), and microsatellite stability (MSS). Variants with potential clinical significance were detected in BRAF (p.Asn581Ile), TP53 (c.375+1G>T), and PTPRT (c.4199_4200delinsT; p.Gln1400LeufsTer11). Resection margins were negative for invasive carcinoma. An unexpected concurrent finding was extensive amyloid deposition in lung parenchyma and multiple sampled lymph node stations (Figure 2D–G). Congo red staining confirmed amyloid, and LC–MS/MS typing identified AL (lambda) amyloid. Gross and histologic features of residual lung adenocarcinoma after neoadjuvant therapy. (A) Representative gross photograph of the right upper lobectomy specimen demonstrating a firm tumour bed corresponding to the treated lesion. (B) Haematoxylin and eosin (H&E) section showing residual adenocarcinoma within therapy‐associated stromal change/treatment effect. (C) TTF‐1 immunohistochemistry highlighting nuclear positivity in residual tumour cells, supporting pulmonary adenocarcinoma lineage. H&E sections demonstrate extensive extracellular, amorphous eosinophilic material within lung parenchyma/vascular or interstitial compartments (D) and lymph nodes (F), morphologically compatible with amyloid. (E, G) Congo red stain highlights corresponding deposits, confirming amyloid. Each staining level is shown at 400× magnification. Given confirmation of an AL subtype, evaluation for a plasma cell dyscrasia was pursued. Bone marrow biopsy in March 2025 showed 6.4% clonal, lambda‐restricted plasma cells. Congo red staining of marrow was negative. Fluorescence in situ hybridisation demonstrated t(11;14) (CCND1/IGH fusion). Serum free light chain testing in March 2025 showed kappa 0.64 mg/L and lambda 25.54 mg/dL with a markedly decreased ratio (0.025). Cardiac evaluation in March 2025 included transthoracic echocardiography, which demonstrated mildly reduced systolic function (3D LVEF 47%), concentric left ventricular hypertrophy (IVSd 1.3 cm), and reduced global longitudinal strain (GLS −9.2%), prompting consideration of possible cardiac involvement. Cardiac MRI (April 2025) was interpreted as non‐ischaemic cardiomyopathy, with no imaging evidence of cardiac amyloidosis. Subsequent specialty evaluation assessed no clinically meaningful cardiac involvement at that time and considered alternative contributors, including prior radiation‐associated cardiac disease and long‐standing conduction abnormalities. The postoperative multidisciplinary review recommended adjuvant immunotherapy, consistent with contemporary NSCLC management. Adjuvant pembrolizumab was administered (six cycles beginning in February 2025 per timeline). For systemic AL amyloidosis, the patient initiated daratumumab in combination with cyclophosphamide, bortezomib, and dexamethasone (daratumumab‐CyBorD) in April 2025 and continued through June 2025. He achieved a partial haematologic response, with serum‐free lambda light chain 82.9 mg/L and dFLC 81 mg/L at the most recent reported assessment. In January 2026, he was diagnosed with bilateral carpal tunnel syndrome, supporting systemic involvement. He remains under longitudinal follow‐up with haematology and cardiology, with ongoing surveillance for NSCLC recurrence. Discussion This case highlights practice‐relevant considerations for thoracic oncology and pathology. Amyloid may be identified incidentally in oncologic resections, including post‐neoadjuvant specimens in which therapy‐associated stromal change can be prominent. Recognition of amorphous extracellular deposits and confirmatory Congo red staining are essential [3]. When amyloid is present, definitive subtyping is critical because downstream evaluation and management differ substantially between AL and AA amyloidosis. LC–MS/MS provided high‐confidence typing in this case and directly informed subsequent diagnostic workup. A key morphologic differential diagnosis for amyloid‐like pulmonary deposits is pulmonary light chain deposition disease (LCDD), which represents nonfibrillar immunoglobulin light chain deposition, and may mimic amyloid on H&E [4]. In lung specimens, Masson trichrome and sulfated Alcian blue have been reported to help distinguish LCDD from amyloidosis when Congo red findings are equivocal [5]. Recognition should prompt targeted clinicopathologic correlation and hematologic evaluation. Reported LCDD case series emphasise that pulmonary involvement is uncommon, can present with variable radiographic patterns, and is frequently linked to an underlying plasma cell dyscrasia. Pulmonary amyloidosis can resemble malignancy on imaging and may only be diagnosed through surgical sampling, emphasising the need for histologic confirmation in FDG‐avid cases. AL (lambda) amyloid prompted evaluation for an underlying plasma cell clone. The marrow findings (lambda restriction with t(11;14)) and marked free light chain imbalance supported systemic AL amyloidosis despite Congo red negativity in marrow. Organ involvement can be heterogeneous and evolve over time. Here, echocardiography and strain findings raised concern, whereas cardiac MRI did not show imaging evidence of amyloid, and specialty review favoured limited clinical cardiac involvement at that time. This discordance supports a multidisciplinary approach incorporating symptoms, biomarkers, echocardiography and advanced imaging. The temporal relationship to neoadjuvant pembrolizumab‐based therapy raises the possibility of association but does not establish causality. Available outside pretreatment pathology and cytology reports did not describe amyloid in the core biopsy or EBUS‐guided nodal specimens, and pretreatment PET/CT showed no FDG‐avid thoracic nodal or distant disease. However, because archival pretreatment tissue and serum were unavailable for additional amyloid‐directed studies, the chronology of the plasma cell dyscrasia relative to neoadjuvant therapy remains uncertain. The amyloidosis may therefore have been pre‐existing but clinically occult, only becoming apparent after surgical resection and subsequent hematologic evaluation. As ICIs are increasingly used in curative‐intent NSCLC, recognition and reporting of unexpected coexisting diagnoses remain important for multidisciplinary care. Systemic AL (lambda) amyloidosis was identified in lung parenchyma and lymph nodes at lobectomy after neoadjuvant pembrolizumab‐based chemoimmunotherapy for resectable lung adenocarcinoma. This case highlights the need to recognise amyloid in thoracic specimens, confirm the subtype using proteomic methods and coordinate hematologic evaluation and organ staging alongside curative‐intent NSCLC care. Author Contributions Weijie Ma: conceptualisation, writing – original draft. Dana M. Ferrari‐Light: writing, reviewing and editing. Liang Lu: conceptualisation, writing – reviewing and editing. Funding The authors have nothing to report. Consent The authors declare that written informed consent was obtained for the publication of this manuscript and accompanying images using the consent form provided by the Journal. Conflicts of Interest The authors declare no conflicts of interest.