ALK-rearranged lung adenocarcinoma transforming to mucoepidermoid carcinoma in a 45-year-old female: a case report Authors: Yachen Liu, Xi Liu, Hongmeng Ma, Wenxin Li Published: 2026 ABSTRACT AbstractBackgroundPulmonary mucoepidermoid carcinoma (PMEC) accounts for less than 1% of pulmonary malignancies and is predominantly primary, exhibiting distinct clinical manifestations and treatment responses compared to lung adenocarcinoma (LUAD). Although the incidence of histological transformation from non-small cell lung cancer (NSCLC) to small cell lung cancer (SCLC) following tyrosine kinase inhibitor (TKI) therapy is approximately 3–14%, transformation from LUAD to PMEC is an extremely rare event. Currently, there is a lack of unified understanding and diagnostic/therapeutic guidelines concerning its mechanism, clinical features, treatment strategies, and prognosis. This article reports a case of HT from LUAD to PMEC following anaplastic lymphoma kinase (ALK)-TKI therapy, aiming to enhance clinicians’ awareness of this rare condition and provide preliminary reference for the diagnosis and management of similar cases.Case DescriptionThis report describes a 45-year-old female patient who was admitted to the hospital with a chief complaint of “persistent cough, sputum production, and dyspnea for 2 years, accompanied by chest tightness for 1 week”. She was diagnosed with ALK exon 20 mutation-positive left LUAD (cT1N1M1, stage IV). After 28 months of first-line ALK-TKI alectinib therapy, the patient experienced disease progression. Repeated pathological biopsy confirmed a transition from LUAD to PMEC, with persistent ALK mutation. Subsequent treatment included lorlatinib and third-line chemotherapy (pemetrexed/carboplatin regimen for four cycles), achieving acceptable disease control. The progression-free survival (PFS1) prior to transformation was 8 months, and the patient is currently receiving palliative care.ConclusionsTumors that undergo LUAD-to-PMEC transformation after ALK-TKI therapy exhibit persistently low levels of ALK fusion and characteristic immunohistochemical alterations [deletion of thyroid transcription factor-1 (TTF-1)/Napsin A, upregulated expression of CK5/6/P40], along with inherent resistance to ALK-TKIs. When heterogeneous resistance emerges during TKI therapy, prompt repeat biopsy with concurrent pathological and next-generation sequencing (NGS) testing is recommended. If confirmed, immediate transition to phenotype-specific therapy should be initiated. Although this is a single-case observation requiring validation in larger cohorts, these findings provide preliminary guidance for clinical identification and management of such rare transformations.Keywords: Case report, lung adenocarcinoma (LUAD), pulmonary mucoepidermoid carcinoma (PMEC), histologic transformation Highlight box Key findings • This first reported case of anaplastic lymphoma kinase (ALK)-positive lung adenocarcinoma (LUAD) transforming into pulmonary mucoepidermoid carcinoma (PMEC) after ALK-tyrosine kinase inhibitor (TKI) therapy reveals a “fusion-retained but functionally independent” state with intrinsic lorlatinib resistance. We recommend early repeat biopsy upon heterogeneous resistance and switching to phenotype-directed therapy if PMEC transformation is confirmed. What is known and what is new? • TKI resistance in ALK-rearranged LUAD can occur via histological transformation (most commonly to small cell lung cancer), with retained ALK fusion and frequent TP53/RB1 loss. • This first reported case of transformation to PMEC shows retained low-abundance ALK fusion without TP53/RB1 mutations, exhibiting an “ALK fusion-retained but functionally independent” state with intrinsic lorlatinib resistance. What is the implication, and what should change now? • ALK-TKI resistance can arise from rare histological transformations (e.g., to PMEC) that retain ALK fusion but become functionally independent, rendering further TKI therapy ineffective. • Upon heterogeneous TKI resistance, perform early repeat biopsy. If rare transformation is confirmed, switch to phenotype-directed treatment (e.g., surgery or chemotherapy) instead of continuing TKIs. Introduction Histologic transformation (HT) is observed in approximately 3–14% of epidermal growth factor receptor (EGFR)-mutated cases and a minority of anaplastic lymphoma kinase (ALK)-rearranged cases (1,2). It represents a common mechanism of acquired resistance to tyrosine kinase inhibitor (TKI) therapy in transformed small cell lung cancer (SCLC), frequently driven by TP53 and RB1 gene inactivation (3,4). Pulmonary mucoepidermoid carcinoma (PMEC) is a rare salivary gland-type carcinoma, accounting for only 0.1–0.2% of pulmonary malignancies (5). Non-small cell lung cancer (NSCLC)-to-PMEC transformation is extremely rare, and its pathogenesis, clinical features, treatment, and prognosis remain unclear, with no established diagnostic or therapeutic guidelines currently available. This report describes a female patient with advanced ALK-mutated lung adenocarcinoma (LUAD) who underwent ALK-TKI therapy and subsequently developed HT into PMEC. Through literature review, we explore the clinical and molecular characteristics of this rare transformation type, aiming to expand the spectrum of HTs associated with TKI-acquired resistance, enhance clinicians’ understanding of such rare cases, and provide reference for the diagnosis and management of similar cases. We present this article in accordance with the CARE reporting checklist (available at https://acr.amegroups.com/article/view/10.21037/acr-2026-0062/rc). Case presentation (Figures 1-4) H&E-stained sections of the patient’s tumor before and after transformation. (A) H&E-stained section, ×5 (pre-transformation); (B) H&E-stained section, ×10 (pre-transformation); (C) H&E-stained section, ×5 (post-transformation); (D) H&E-stained section, ×10 (post-transformation). H&E, hematoxylin and eosin. Serial chest CT imaging during the disease course. Panels A1 and A2 show the baseline target lesion. Panels B1 through H2 display follow-up scans obtained on February 10, 2023; January 11, 2024; June 22, 2024; September 26, 2024; March 14, 2025; and June 6, 2025, respectively. Panels D and E show identical images; panel E is duplicated only for optimized layout arrangement. CT, computed tomography. Serial abdominal CT imaging during the disease course. Panels A through F correspond to follow-up scans obtained on April 12, 2024; June 22, 2024; September 26, 2024; January 10, 2025; March 14, 2025; and June 6, 2025, respectively. CT, computed tomography. Summary of the patient’s treatment. m, months. A 45-year-old female was admitted on May 13, 2022, due to “cough, sputum, and shortness of breath for over 2 years, accompanied by chest tightness for 1 week”. Physical examination revealed coarse breath sounds in both lungs, dry rales in the left lung, and no palpable enlargement of superficial lymph nodes. There was no relevant medical history, smoking or alcohol consumption history, or family history of cancer. Post-admission pathological biopsy, immunohistochemistry (IHC), and morphological examination revealed positivity for cytokeratin (CK), cytokeratin 7 (CK7), Napsin A, thyroid transcription factor-1 (TTF-1), and focal mucin 2 (MUC-2), with a Ki-67 proliferation index of 3–5%, consistent with LUAD (Figure 1). Baseline tumor tissue next-generation sequencing (NGS) detected an ALK fusion mutation, though specific variant allele frequency (VAF) values were not provided. The diagnosis was confirmed as left LUAD (cT1N1M1, stage IV), with metastases to the left hilar lymph nodes and the ninth thoracic vertebra (T9). During the waiting period for genetic testing, empirical chemotherapy was administered for one cycle (carboplatin 600 mg + pemetrexed 750 mg). Subsequently, a treatment regimen of alectinib (600 mg bid) and desmopressin (120 mg q28d) was initiated, followed by disease progression after 28 months. The progressive physical examination revealed coarse breath sounds in both lungs, no dry or moist rales, and no palpable enlargement of superficial lymph nodes. Re-biopsy IHC results indicated: CK5/6 (+), CK7 (+), p63 (+), p40 (+), wild-type p53 (30–40%, with variable expression intensity), CgA (−), Syn (−), TTF-1 (−), Napsin A (−), CR (−), and Ki-67 (+) (3–8%). Inner Mongolia People’s Hospital’s pathology team suspected well-differentiated squamous cell carcinoma, possibly with PMEC, and required multicenter consultation for exclusion (Figure 1). Subsequent inter-institutional consultation confirmed PMEC, with NGS detecting an ALK EML4-20: ALK-20 fusion mutation and a VAF of 3.00%. An ALK EML4-20: ALK-20 fusion was identified with a VAF/copy number of 3.00%. Programmed death-ligand 1 (PD-L1) testing revealed a tumor proportion score (TPS) of 5% and a comprehensive positive score (CPS) of 10. The diagnosis was revised to left PMEC (cT4NxM1, stage IV), involving the left pulmonary arteries and veins, metastases to the left hilar lymph nodes, right 4th rib, 9th thoracic vertebra, and liver. As well as the right lower lobe, the left cervical and supraclavicular lymph nodes pending exclusion. The partial follow-up results from the initial diagnosis to the second biopsy stage of disease progression are as follows (based on RECIST 1.1 criteria): Baseline period (May 16, 2022): chest contrast-enhanced computed tomography (CT) revealed obstruction of the left lower lobe and dorsal segment bronchi, with a hilar mass measuring approximately 30 mm × 26 mm. Baseline period (May 16, 2022): chest contrast-enhanced computed tomography (CT) revealed obstruction of the left lower lobe and dorsal segment bronchi, with a hilar mass measuring approximately 30 mm × 26 mm. July 31, 2022: the nodule in the dorsal segment of the lower left lung lobe shrank to 26 mm × 10 mm (versus baseline), indicating a partial response (PR). July 31, 2022: the nodule in the dorsal segment of the lower left lung lobe shrank to 26 mm × 10 mm (versus baseline), indicating a partial response (PR). February 10, 2023: a nodule measuring approximately 17 mm × 12 mm was identified in the dorsal segment of the lower left lung lobe, with dimensions largely consistent with previous findings. February 10, 2023: a nodule measuring approximately 17 mm × 12 mm was identified in the dorsal segment of the lower left lung lobe, with dimensions largely consistent with previous findings. January 11, 2024: a soft tissue density nodule in the left hilar region near the dorsal segment of the lower left lung lobe, measuring approximately 23 mm × 19 mm, with size comparable to previous findings. January 11, 2024: a soft tissue density nodule in the left hilar region near the dorsal segment of the lower left lung lobe, measuring approximately 23 mm × 19 mm, with size comparable to previous findings. April 12, 2024: the nodule in the dorsal segment of the left lower lobe increased from 23 mm × 19 mm to 25 mm × 21 mm. Abdominal CT and brain magnetic resonance imaging (MRI) showed no metastasis, and the condition was assessed as stable disease (SD). April 12, 2024: the nodule in the dorsal segment of the left lower lobe increased from 23 mm × 19 mm to 25 mm × 21 mm. Abdominal CT and brain magnetic resonance imaging (MRI) showed no metastasis, and the condition was assessed as stable disease (SD). June 22, 2024: enhanced chest CT revealed that the nodule in the dorsal segment of the left lower lobe had enlarged to 35 mm × 22 mm with blurred margins, involving the left main pulmonary artery and left lower lobar pulmonary artery. A new 16 mm × 14 mm subpleural nodule in the left upper lobe (posterior apical segment) suggested metastasis. Enhanced abdominal CT demonstrated a new hypoperfusion hypodense lesion beneath the right hepatic capsule, with a maximum diameter of 39 mm. The condition was assessed as progressive disease (PD). The patient and family insisted on continuing the original ALK-TKI therapy. June 22, 2024: enhanced chest CT revealed that the nodule in the dorsal segment of the left lower lobe had enlarged to 35 mm × 22 mm with blurred margins, involving the left main pulmonary artery and left lower lobar pulmonary artery. A new 16 mm × 14 mm subpleural nodule in the left upper lobe (posterior apical segment) suggested metastasis. Enhanced abdominal CT demonstrated a new hypoperfusion hypodense lesion beneath the right hepatic capsule, with a maximum diameter of 39 mm. The condition was assessed as progressive disease (PD). The patient and family insisted on continuing the original ALK-TKI therapy. September 26, 2024: the nodule in the dorsal segment of the lower left lung lobe further enlarged to 37 mm × 38 mm with ill-defined borders. The nodule in the upper left lung lobe shrank to 15 mm × 8 mm. The hepatic lesion increased in size to 42 mm. This was assessed as PD. The patient was re-admitted on October 13, 2024 (Figure 2, A1-F2; Figure 3A-3C). September 26, 2024: the nodule in the dorsal segment of the lower left lung lobe further enlarged to 37 mm × 38 mm with ill-defined borders. The nodule in the upper left lung lobe shrank to 15 mm × 8 mm. The hepatic lesion increased in size to 42 mm. This was assessed as PD. The patient was re-admitted on October 13, 2024 (Figure 2, A1-F2; Figure 3A-3C). Following a secondary biopsy, a change in treatment to chemotherapy was recommended, but the patient and family insisted on oral third-generation ALK-TKI therapy (lorlatinib 100 mg once daily). After four months of treatment, the patient was readmitted due to “2 months of persistent dyspnea during targeted therapy for LUAD”. After repeated discussions, the patient agreed to terminate the targeted therapy and initiate systemic chemotherapy. Following four cycles of third-line chemotherapy (pemetrexed 800 mg + carboplatin 500 mg), the patient experienced anemia and worsening systemic condition due to vaginal bleeding recurrence. After deliberation, treatment was temporarily suspended, and the patient subsequently received palliative care at a local hospital. Figure 4 illustrates the patient’s clinical course and therapeutic interventions. The partial follow-up results of the patient after secondary biopsy are as follows: During lorlatinib treatment: chest and abdominal CT scans performed on January 10 and March 14, 2025, both demonstrated continued PD (January 10, 2025: bronchial occlusion in the upper and lower lobes of the left lung, extensive pulmonary consolidation and atelectasis, increased left pleural effusion. Maximum diameter of the lesion in the right hepatic lobe was approximately 45 mm. March 14, 2025: new right pleural effusion) (Figure 2, G1,G2; Figure 3D,3E). During lorlatinib treatment: chest and abdominal CT scans performed on January 10 and March 14, 2025, both demonstrated continued PD (January 10, 2025: bronchial occlusion in the upper and lower lobes of the left lung, extensive pulmonary consolidation and atelectasis, increased left pleural effusion. Maximum diameter of the lesion in the right hepatic lobe was approximately 45 mm. March 14, 2025: new right pleural effusion) (Figure 2, G1,G2; Figure 3D,3E). After four cycles of chemotherapy: on June 6, 2025, enhanced CT scans of the chest and abdomen revealed bronchial occlusion with atelectasis in both upper and lower lobes of the left lung, partial re-expansion of the left lower lobe compared to March 14, 2025, and bilateral pleural effusion. The subcapsular liver lesion measured 25 mm × 35 mm and exhibited a fuller morphology than previously (Figure 2, H1,H2; Figure 3F). After four cycles of chemotherapy: on June 6, 2025, enhanced CT scans of the chest and abdomen revealed bronchial occlusion with atelectasis in both upper and lower lobes of the left lung, partial re-expansion of the left lower lobe compared to March 14, 2025, and bilateral pleural effusion. The subcapsular liver lesion measured 25 mm × 35 mm and exhibited a fuller morphology than previously (Figure 2, H1,H2; Figure 3F). All procedures performed in this study were in accordance with the ethical standards of the institutional and/or national research committee(s) and with the Declaration of Helsinki and its subsequent amendments. Written informed consent for publication of this case report and accompanying images was not obtained from the patient or the relatives after all possible attempts were made. Discussion Mucoepidermoid carcinoma (MEC) is typically a salivary gland tumor, while PMEC is extremely rare. PMEC is classified into low- and high-grade based on cellular composition and pleomorphism, and into central and peripheral types based on anatomical location (6,7). Central-type PMEC is the most common. Among them, low-grade PMEC often presents as a well-defined central endobronchial mass, whereas high-grade PMEC tends to present as a poorly defined peripheral lesion. Low-grade PMEC have a favorable prognosis, whereas high-grade PMEC exhibit greater invasiveness (8,9). Defined by a combination of mucin-secreting, squamous, and intermediate cells (10,11), PMEC often presents with bronchial irritative symptoms such as cough, expectoration, hemoptysis, chest pain, chest tightness, and fever (12). IHC, cytochemistry, and MAML2 translocation analysis aid in the diagnosis of PMEC. IHC of PMEC typically shows positivity for CK5/6, CK7, p63, and p40, negativity for CK20 and SMA, a Ki-67 index ranging from 5% to 60%, and negativity for LUAD markers TTF-1 and Napsin A (13,14). No typical driver gene mutations characteristic of lung cancer had been identified in PMEC. While MAML2 gene rearrangement is a defining molecular feature of primary PMEC (15), it is relatively rare in high-grade PMEC and has never been detected in LUAD (16). Given that low-grade PMEC morphologically resembles solid LUAD, the diagnosis of transformation from LUAD to PMEC poses a significant challenge. Therefore, pathologists should rely on a combination of cellular components and IHC markers to distinguish PMEC from solid LUAD. Although MAML2 rearrangement testing can be a helpful adjunct in diagnosing LUAD-to-PMEC transformation, it is not considered essential, as studies have shown that approximately 20–50% of PMEC cases (particularly high-grade tumors) are negative for MAML2 rearrangement, and diagnosis can be reliably established based on morphology and IHC alone (13,15). Here, the initial biopsy of the patient showed LUAD with columnar tumor cells arranged in glandular structures and abundant cytoplasm. IHC was positive for CK, CK7, Napsin A, and TTF-1, consistent with typical LUAD. The re-biopsy specimen displayed solid nests of mucinous, squamous, and intermediate cells, consistent with PMEC histology (17). IHC comparison showed Syn (−), CgA (−), CK7 (+), and Ki-67 (3–5% pre- vs. 3–8% post-transformation). However, TTF-1 and Napsin A became negative, while CK5/6 and P40 turned positive post-transformation—aligning with known PMEC immune-profile (18). The repeat biopsy of the patient followed by genomic analysis identified an EML4-ALK fusion mutation, with no mutations detected in other common driver genes such as EGFR, TP53, RB1, PIK3CA, or MET. Due to the economic constraints, MAML2 gene rearrangement testing was not performed on the biopsy samples during the conversion period. Combined with genetic findings, this confirms HT from ALK-rearranged LUAD to PMEC after alectinib resistance, rather than NSCLC with MEC differentiation (19). The potential mechanisms underlying pathological transformation include tumor heterogeneity, acquired resistance (such as secondary mutations, pathway activation, or HT), and selective pressure from targeted therapies. HT is one of the key mechanisms underlying resistance to TKIs in NSCLC harboring driver genes (such as EGFR mutations, ALK rearrangements, and ROS1 mutations), with the highest transformation rate observed in EGFR-mutated cases (3–14%) (4,20,21). Transformed tumors typically retain the original driver gene, which serves as a critical distinguishing feature between transformed and de novo tumors (22). For example, LUAD with EGFR mutations retains its initial EGFR mutation during transformation into SCLC or lung squamous cell carcinoma (LUSC) (23,24). In LUAD with ALK rearrangement, the ALK fusion persists during its transformation into SCLC (22,25). It is noteworthy that during the transformation process, tumors frequently acquire additional genetic mutations while retaining their original driver genes. In the case of NSCLC to SCLC transformation, the inactivation of TP53 and RB1 constitutes a pivotal driving event. However, genetic testing in this ALK-rearranged LUAD case that transformed to PMEC identified only low-abundance ALK fusions. No TP53 or RB1 mutations or other common driver genes alterations, such as EGFR, PIK3CA or MET, were detected. This suggests that during transformation from LUAD to PMEC, the original driver genes detected in the LUAD may be retained, whereas the presence of additional or alternative driver mutations remains uncertain. According to the following characteristics, we hypothesize the transformation mechanism is more consistent with the clonal selection model (26,27). Firstly, the persistent ALK fusion was detected both before and after transformation suggesting that PMEC and adenocarcinoma may share a common origin. Secondly, there were no other driver gene mutations (e.g., TP53/RB1 inactivation) were identified. Given that EGFR, KRAS, BRAF, ALK, PIK3CA, PDGFRA, and DDR2 are reportedly not typical driver genes in primary PMEC, this implies that transformed PMEC may have distinct molecular features from its primary counterpart (28). These results suggested that under TKI pressure, a subclone with lower ALK dependence may be selected, driving PMEC transformation in the absence of novel driver mutations. Although the ALK fusion is retained following transformation to PMEC, the tumor exhibits ALK-independent growth, ultimately rendering even potent TKIs ineffective. This explains the suboptimal efficacy of lorlatinib. Additionally, long-term ALK-TKI selective pressure may have facilitated the transformation process, suggesting the potential existence of unidentified molecular mechanisms. This case reveals a distinct transformation subtype that retains ALK fusion but lacks acquired driver mutations and exhibits ALK-independent growth, sharply contrasting with classic SCLC transformation. It should be noted that the biological behavior of transformed PMEC may differ from that of primary PMEC. Literature reports indicate that surgical resection is the standard treatment for primary PMEC, with 3-, 5-, and 10-year survival rates of 94%, 88%, and 88%, respectively. Low-grade PMEC has a favorable prognosis, whereas high-grade PMEC has a poorer prognosis, with a postoperative recurrence rate of 25% (29). For unresectable primary PMEC, there is currently a lack of chemotherapy data and guidelines. Additionally, studies have demonstrated that ALK-mutated primary PMEC respond well to lorlatinib therapy, suggesting that EML4-ALK may serve as a potential target for primary PMEC (30). However, the resistance to lorlatinib observed in the transformed PMEC in this case differs from that of primary PMEC, a discrepancy that may be attributable to alterations in tumor biological characteristics during the transformation process. Consequently, although targeted therapy may benefit primary PMEC with identified driver mutations, it offers limited efficacy for transformed PMEC. Clinically, we suggested that after histological transformation, when heterogeneous TKI resistance occurs, early repeat biopsy with concurrent pathological and NGS testing is recommended. If rare transformation (e.g., PMEC) is confirmed, treatment should shift to phenotype-directed therapy (e.g., surgery or platinum-based chemotherapy) rather than continuing the original TKI. The molecular mechanisms underlying these differences warrant further investigation in future studies. This study has the following limitations. First, the original imaging reports did not provide details on the morphological evolution of the lesions, only recording quantitative data and thus could not assess whether PMEC transformation was accompanied by characteristic imaging changes. Second, the baseline reports did not provide VAF values, making quantitative comparison impossible. Only qualitative inferences could be drawn regarding whether the tumor had potentially transitioned to ALK-independent growth post-transformation. Third, due to financial constraints, MAML2 gene rearrangement testing was not performed. Therefore, whether the transformed PMEC in this case shares the same molecular driver mechanism as primary PMEC remains unclear. Fourth, this case represents a retrospective single-case analysis, and conclusions require validation with additional cases. Future studies should, where resources permit, conduct detailed imaging assessments of suspected transformation cases, standardize VAF reporting, routinely detect MAML2 status, and perform functional experiments to validate the clonal selection hypothesis. This case demonstrates that PMEC transformation following ALK-TKI therapy can manifest as a unique state characterized by retained ALK fusion but functional independence. It also provides clinical guidance that early repeat biopsy should be performed in cases of heterogeneous resistance, and if PMEC transformation is confirmed, phenotype-targeted therapy should be initiated instead of continuing TKI therapy. Future multicenter case studies are required to validate this finding. Conclusions This case represents the first report of PMEC transformation in LUAD following ALK-TKI therapy, revealing a unique resistance state characterized by preservation of ALK fusion but functional independence. This finding expands the spectrum of HT associated with TKI resistance and suggests that when TKI-induced heterogeneous resistance occurs, repeat biopsy should be actively performed, and after transformation, a phenotype-targeted treatment approach should be adopted. Although this is a single-case observation requiring further validation, these conclusions provide preliminary guidance for the clinical identification and management of such rare transformations. Supplementary The article’s supplementary files as Acknowledgments None. Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. All procedures performed in this study were in accordance with the ethical standards of the institutional and/or national research committee(s) and with the Declaration of Helsinki and its subsequent amendments. Written informed consent for publication of this case report and accompanying images was not obtained from the patient or the relatives after all possible attempts were made. Footnotes Reporting Checklist: The authors have completed the CARE reporting checklist. Available at https://acr.amegroups.com/article/view/10.21037/acr-2026-0062/rc Funding: This work was supported by the Beijing Cancer Prevention and Treatment Research Association (No. KY202501010). Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://acr.amegroups.com/article/view/10.21037/acr-2026-0062/coif). All authors report the funding from the Beijing Cancer Prevention and Treatment Research Association (No. KY202501010). The authors have no other conflicts of interest to declare. References Available at https://acr.amegroups.com/article/view/10.21037/acr-2026-0062/prf Supplementary Materials The article’s supplementary files as