Figure 3. Baseline 18F-FDG PET/CT, fused images. Axial slices at the level of the nasal vestibule and of the oropharynx show intense uptake within the left naso-labial mass (SUVmax 11.26; MTV40 28 cm³); the sagittal reformats (bottom row) show anterior and inferior extension towards the upper lip and the hard palate, with no hypermetabolic nodal or visceral disease elsewhere.

Discussion

A lymphoma concealed by a “benign” midline disease

The most instructive feature of this case is the length and the nature of the pre-diagnostic pathway. For close to fifteen years the destructive midfacial process was attributed to chronic maxillo-mandibular osteitis, prompting two operations, and subsequently to symptomatic septal deviation. Such trajectories are frequently described in ENKTL-NT: non-specific symptoms precede ulceration and local destruction by months to years, and the disease is commonly labelled chronic rhinosinusitis, fungal sinusitis, odontogenic osteomyelitis or granulomatosis with polyangiitis [8–10]. Andreou et al. emphasise that oral warning signs — palatal involvement, gingival ulceration, tooth mobility and exfoliation, and oral malodour — are often overlooked [8]. In our patient the oro-nasal communication, the rhizalysis of teeth 21 and 23 and the naso-genian soft-tissue infiltration already visible on the May 2025 CT are, in retrospect, entirely compatible with tumour infiltration, which illustrates how difficult it is to separate a chronic infective process from a lymphomatous one on imaging alone.

This confusion is not anecdotal. Maghrebian and Mediterranean series report diagnostic delays of several months, usually after repeated courses of antibiotics or anti-inflammatory drugs and, in some cases, after a first non-diagnostic biopsy [4–7]. In a series of 15 patients, Tababi et al. stressed the need to repeat sampling in any destructive sinonasal lesion of undetermined cause [6].

Presentation after septal surgery

The postoperative onset deserves comment. The vestibular swelling appeared in the immediate aftermath of septoplasty with turbinoplasty, which initially suggested an infective or inflammatory complication of the procedure. Two non-exclusive hypotheses may be advanced: that the disease was already present but subclinical at the time of surgery and simply continued its natural course; or that tumour growth was accelerated by surgical trauma and the accompanying local inflammatory response. Either way, this observation argues for the routine histopathological examination of every turbinectomy and septal resection specimen — particularly when the mucosa appears macroscopically unusual — and for prompt re-evaluation of any postoperative swelling with an atypical course. In our patient the surgical specimens were discarded, and an opportunity for earlier diagnosis was probably lost.

Histological pitfalls and the role of EBER

The histological diagnosis of ENKTL-NT is notoriously difficult. Extensive ischaemic necrosis related to angiodestruction, frequent crush artefact in small specimens and an abundant reactive inflammatory infiltrate account for the high proportion of non-diagnostic biopsies [8,10]. In our case the proliferation was described as “largely crushed”, and the diagnosis rested on the few preserved areas. Deep, multiple, adequately sized biopsies taken from non-necrotic tissue are therefore essential.

The immunophenotype observed here — CD45+, CD56+, granzyme B+, focal CD3+, CD20−, AE1/AE3− — is characteristic. CD3 positivity on paraffin sections usually reflects cytoplasmic CD3ε, the surface chain being absent from NK cells; negative CD20 and negative cytokeratins exclude B-cell lymphoma and undifferentiated carcinoma respectively, the principal differential diagnoses at this site [1,3]. Demonstration of EBV by EBER in situ hybridisation nevertheless remains the reference criterion required by the 2022 WHO classification, and its absence is the main limitation of our report [1,3]. Access to this technique should be regarded as a priority in any centre managing sinonasal tumours.

Staging and the contribution of metabolic imaging

18F-FDG PET/CT is now the reference investigation for initial staging and response assessment in ENKTL-NT, whose lesions are consistently FDG-avid [3,11]. In our patient it delineated local extension — cartilaginous septum, limen nasi, inferior turbinate, nasolacrimal duct and palate — and, more importantly, established that the disease was strictly localised, with no nodal or visceral hypermetabolism. This directly determined management, since radiotherapy is central to the treatment of localised disease [11,17]. The prognostic value of baseline metabolic parameters, in particular SUVmax and metabolic tumour volume, has been reported by several groups, with higher SUVmax associated with shorter survival, although these data remain heterogeneous and no consensus threshold exists [3].

Prognostic assessment

The outlook in ENKTL-NT was long regarded as dismal, with median survival under one year in older series treated with CHOP-like regimens, largely because tumour cells express P-glycoprotein and are therefore resistant to anthracyclines [3,11]. Several prognostic models have been proposed: the Korean model of Lee et al., combining B symptoms, stage, lactate dehydrogenase and regional nodal involvement [13]; and the PINK score, developed in the era of non-anthracycline-based therapy and based on age over 60 years, stage III/IV disease, distant nodal involvement and extranasal location [12]. Extranasal disease in particular carries a significantly worse prognosis than the classical nasal form [14]. Our patient combined favourable features — age under 60, stage IE, nasal location and no nodal involvement — giving a PINK score of 0 and low-risk classification, consistent with the complete remission achieved.

Treatment

Treatment of localised disease now rests on non-anthracycline chemotherapy containing L-asparaginase or pegaspargase combined with high-dose involved-field radiotherapy, generally 50–56 Gy [11,17,18]. Three combination strategies have been evaluated: radiotherapy followed by chemotherapy, concurrent chemoradiotherapy, and the so-called sandwich approach in which radiotherapy is bracketed by chemotherapy cycles [11]. The GELOX regimen (gemcitabine, L-asparaginase, oxaliplatin) sandwiched with 56 Gy achieved a complete response rate of 74% and an overall response rate of 96%, with 5-year overall survival of 85% in the updated analysis [11,15]. Substituting pegaspargase for L-asparaginase (P-GEMOX) simplifies administration and improves immunological tolerance [11]. Concurrent chemoradiotherapy combining 50 Gy with pegaspargase achieved a 90% complete response rate at the end of the concurrent phase and 100% at the end of treatment in a two-centre phase II study [16]. Kwong et al. showed that sequential and concurrent strategies yielded comparable survival [17].

In our patient, three cycles of chemotherapy followed by radiotherapy conform to this combined-modality principle and produced complete clinical and metabolic remission. Surveillance should be intensive during the first two years, when most relapses occur, and should combine clinical examination, nasal endoscopy, metabolic imaging and — where available — quantification of circulating EBV DNA, whose kinetics provide a sensitive marker of tumour burden and of relapse [3,18].

Conclusions

ENKTL-NT remains a difficult diagnosis, and delay is the rule rather than the exception. Our case illustrates a particularly misleading presentation, in which a destructive midfacial lesion was managed for more than a decade as chronic maxillo-mandibular osteitis, the diagnosis being reached only when a vestibular swelling appeared after septoplasty. Any chronic, destructive or ulcero-proliferative midline facial lesion that fails to respond to conventional treatment should lead to deep and repeated biopsies with complete immunohistochemistry and testing for EBV, and turbinate and septal specimens should be submitted routinely for histopathology. Staging with PET/CT and the combination of asparaginase-based chemotherapy with involved-field radiotherapy now yield high complete-response rates in localised disease, as illustrated by the favourable outcome in our patient.

Declarations

Ethics approval and consent to participate. [Complete according to journal requirements.]

Consent for publication. Written informed consent was obtained from the patient for publication of this case report and of the accompanying clinical and radiological images. All identifying data have been removed. A copy of the consent form is available for review by the Editor-in-Chief of this journal.

Availability of data and materials. All data generated or analysed during this study are included in this published article.

Competing interests. The authors declare that they have no competing interests.

Funding. No specific funding was received for this work.

Authors' contributions. IG managed the patient, collected the data and drafted the manuscript. RC and SB participated in the management of the patient and in data collection. ML, ML, WB, YO, SR, RLA and MR critically revised the manuscript for important intellectual content. MM supervised the work and gave final approval. All authors read and approved the final manuscript.

Acknowledgements. The authors thank the Departments of Pathology, Radiology, Nuclear Medicine and Haematology–Oncology of Ibn Rochd University Hospital for their collaboration.