Carcinoma ex pleomorphic adenoma (CXPA) is a rare tumour, with different prevalence rates reported among studies. Epidemiological studies of large series of CXPAs in developing countries are scarce. The aim of the present study was to describe Brazilian patients with CXPA; this was a retrospective study of 38 patients. Demographic and clinico-pathological features were evaluated. No preferential gender was found, and the mean age at diagnosis was 57.6 years. The most commonly involved site was the parotid, followed by the submandibular and the minor salivary glands. A prevalence of clinical stages III and IV was observed at diagnosis. The most common histological subtypes were salivary duct carcinoma, adenocarcinoma not otherwise specified, myoepithelial carcinoma, and epithelial–myoepithelial carcinoma. Moreover, by invasive phase, most were frankly invasive carcinoma. Recurrence was observed in seven out of 24 patients with outcome information available, and all were invasive cases. All seven patients died of causes related to the disease. The distributions of cases according to age, gender, tumour location, and clinical stage were similar to those reported in the literature. Frankly invasive cases presented a worse prognosis. More information is needed to further our understanding of the clinico-pathological aspects of CXPA.
Carcinoma ex pleomorphic adenoma (CXPA) is a malignant tumour of the salivary gland. These tumours comprise 3.6% of all salivary tumours, 6.2% of all mixed tumours, and 11.6% of all salivary malignancies. CXPA is uncommon, with a prevalence rate of 5.6 cases per 100,000 malignant neoplasms and an incidence rate of 0.17 tumours per one million persons. It is considered a malignant transformation of a pre-existing pleomorphic adenoma, resulting from the accumulation of tumour-associated genetic alterations to normal cells. Other investigators have suggested that these tumours develop as de novo malignancies, implying that the malignant focus existed from the start of tumour development. Approximately 1.3–7.5% of pleomorphic adenomas can present malignant changes over their natural course. Longevity and recurrence appear to increase the risk of malignant transformation.
The most common site of origin is the parotid gland (67%), but CXPA may originate from the submandibular gland (15%), the sublingual gland (>1%), and minor salivary gland sites (18%), mainly in the soft and hard palate. Although rare, CXPA cases have been described in the sublingual gland, the lacrimal gland, breast, trachea, and nasal cavity. CXPA usually presents in the sixth or seventh decades of life, with a female predominance.
The tumour histopathology comprises a variable proportion of pleomorphic adenoma versus carcinoma. The carcinoma comprises more than 50% of the mass in most cases. In cases without a benign portion, the previous clinico-pathological documentation of pleomorphic adenoma at the same site as the carcinoma must exist. Most frequently, only one histological type is encountered in CXPA, but more than one carcinoma subtype may be present in some cases. The malignant component can be a specific carcinoma type, such as salivary duct carcinoma (SDC), adenocarcinoma not otherwise specified (AdNOS), epidermoid carcinoma (EC), myoepithelial carcinoma (MC), epithelial–myoepithelial carcinoma (EMEC), mucoepidermoid carcinoma (MEC), adenoid cystic carcinoma (ACC), adenosquamous carcinoma (ASC), clear cell adenocarcinoma (CCA), acinic cell adenocarcinoma (ACA), or sarcomatoid carcinoma (SC). The CXPA may be sub-classified as intracapsular or non-invasive, minimally invasive (≤1.5 mm penetration of the malignant component into extracapsular tissue), or frankly invasive (>1.5 mm of invasion from the tumour capsule into the adjacent tissue). The main treatment recommended for CXPA is wide local excision. Adjuvant radiation therapy can be used for cases of invasive tumours, or those with positive margins or vascular or neural invasion.
The reported incidence of CXPA tumours is generally low and varies considerably between series. The largest series previously reported include: 57 cases described by Foote and Frazell, 29 by Beahrs et al., 32 by Moberger and Eneroth, 25 by Gerughty et al., 47 by LiVolsi and Perzin, 146 by Spiro et al., 40 by Tortoledo et al., 102 by Eveson and Cawson, 326 by Gnepp and Wenig, 73 by Lewis et al., 24 by Zbären et al., and 43 by Katabi et al. ; all of these involved North American and European populations. Hence, the aim of this study was to describe a developing country retrospective clinico-pathological analysis involving 38 patients diagnosed with CXPA.
Materials and methods
A retrospective review of the medical records was performed, and 38 Brazilian patients with salivary gland carcinomas diagnosed as CXPA were identified. This study was approved by the ethics committees of the institutions involved. The medical records of patients were examined for demographic data (age and gender) and clinical information (duration of symptoms, medical history, and clinical stage). TNM restaging (T, extent of the tumour; N, spread to lymph nodes; M, metastasis) was reviewed in accordance with the International Union Against Cancer criteria (UICC, 2002). Treatment, recurrence, and follow-up data were also recovered from the medical records.
The histology of the 38 cases of CXPA was reviewed using 5-μm haematoxylin–eosin (H&E) stained sections of formalin-fixed paraffin-embedded samples. The histopathological classification of CXPA was performed in accordance with the World Health Organization classification (WHO, 2005), and all of the cases were sub-classified according to invasiveness as intracapsular, minimally invasive (≤1.5 mm through capsule), or frankly invasive (>1.5 mm through capsule) (Gnepp et al.). Microscopic examination of pleomorphic adenoma showed a well-circumscribed lesion, with intense hyalinization or myxoid areas, clusters and nests of epithelial cells, and ductal structures composed of epithelial and myoepithelial cells. There was no pleomorphism, cellular atypia, or mitosis. On the other hand, the microscopic examination of CXPA showed pleomorphic cells with increased nucleus/cytoplasm ratio, hyperchromatic nuclei and prominent nucleoli, and invasiveness (intracapsular and early and frankly invasive phases). The reaction with Ki-67 antibody was performed in all cases to aid in the definition of the malignant area, and other appropriate immunohistochemistry markers were used to aid in the histopathological sub-classification, particularly in cases of uncertainty.
The studied population consisted of 38 patients with CXPA, of whom 19 were male and 19 were female, with a mean age of 57.6 years (range 27–88 years). The mean period of complaints was 115 months (range 2–360 months), and three (7.9%) had a malignant tumour as a recurrence of pleomorphic adenoma, after 5, 11, and 16 years. The most common symptom was tumour growth (35 cases, 92.1%). Three patients complained of pain, one reported facial paralysis, one headache, one otalgia, and one had salivary flow alterations. Most tumours involved the parotid (28 cases, 73.7%), followed by the submandibular gland (six cases, 15.8%), minor salivary glands of the oral cavity (two palate, one upper lip, 7.8%), and one involved the minor mucous gland from the nasal cavity (2.6%).
The mean tumour size was 5 cm (range 1.5–9.5 cm), with six patients (15.8%) showing invasion of the adjacent structures such as skin, bone, and muscle. At diagnosis, 17 patients (44.7%) presented tumours of an advanced size (T3 or T4 tumours), 11 cases (28.9%) were classified as T1 or T2, and 10 (26.3%) had no information available (and it was not possible to obtain this information). Only six cases (15.8%) had lymph node metastasis at presentation, and one patient (2.6%) had distant metastasis involving the lungs. The final clinical stage was classified as I or II in nine cases (23.7%) and as III or IV in 19 cases (50%); there was no information for 10 cases (26.3%). The clinical parameters of the 38 cases with CXPA are described in Table 1 .
|Age (years)||≤50||10 (26.3)|
|Information not available||4 (10.5)|
|Disease history||Primary||35 (92.1)|
|Recurrent PA||3 (7.9)|
|Submandibular gland||6 (15.8)|
|Minor salivary gland *||4 (10.5)|
|Tumour size||T1||4 (10.5)|
|Information not available||10 (26.3)|
|Clinical stage||I||4 (10.5)|
|Information not available||10 (26.3)|
The most common histological type was SDC (16 cases, 42.1%), followed by AdNOS (eight cases, 21.1%), MC (seven cases, 18.4%), EMEC (five cases, 13.2%), EC (one case, 2.6%), and SC (one case, 2.6%). The invasiveness of the CXPA was subdivided into intracapsular (without invasion), minimally invasive (≤1.5 mm), and frankly invasive (>1.5 mm). Six cases (15.8%) were classified as intracapsular, nine (23.7%) as minimally invasive, and 23 (60.5%) as in the frankly invasive phase. Some of the histopathological classifications and sub-classifications are illustrated in Figs. 1–4 .
Patients were primarily treated with surgery (30 cases, 78.9%); there was no information for eight patients (21.1%). Adjuvant radiotherapy was performed in seven cases (18.4%) due to positive surgical margins. No patient was treated with chemotherapy. Eight patients (21.1%) underwent neck dissection due to positive clinical palpation (N+) and five of these patients (62.5%) had pathologically positive lymph nodes (pN+).
Seven patients (18.4%) experienced tumour recurrences during the follow-up period (from 1 to 80 months) and all were frankly invasive CXPA. Of these seven patients, one (2.6%) had local recurrence, one (2.6%) had regional recurrence, two (5.3%) had distant metastasis (lung and another site without specification), one (2.6%) had local and regional recurrences, one (2.6%) had local and distant (orbital site) recurrences, and one (2.6%) had recurrence without site specification. A total of 15 patients (39.5%) were lost during follow-up and there was no documentation of recurrences or current medical status for these patients. The mean follow-up time was 29 months, ranging from 1 to 115 months. During the follow-up period, 11 patients (28.9%) were alive without evidence of disease, seven patients (18.4%) died of causes related to the disease, and six patients (15.8%) died of causes not related to the disease. Some correlations are described in Table 2 , and the cases are summarized in Table 3 .
|Minor salivary gland||0||1||2||1||0||0||4|
|Clinical status||Alive without disease||2||4||3||1||0||1||11|
|Died of disease||3||1||1||1||1||0||7|
|Died of other causes||4||1||0||1||0||0||6|
|Age (years)||Gender||Site||Time duration||pT (cm) *||T †||N †||M †||Clinical stage||Histopathological subtype||Type||Treatment||Recurrence||Current time||Follow-up (months)|
|61||F||Parotid||20 years||5||T3||N0||M0||III||MC||IC||Surgery, RT||No||AWD||36|
|88||F||SMG||4 months||2||T1||N0||M0||I||SDC||MI||Surgery, RT||No||DOC||45|
|45||M||Parotid||4 years||6||T3||N0||M0||III||SDC||FI||Surgery||Local, RM||DOD||85|
|50||F||Parotid||20 years||8||T3||N1||M0||III||EMEC||FI||Surgery, ND (pN−)|
|74||F||MSG||30 years||4||T2||N0||M0||II||MC||FI||Surgery||Local, DM (eyes)||DOD||24|
|65||M||Parotid||2 years||8||T4a||N1||M0||IVa||SDC||FI||Surgery, ND (pN+)||No||AWD||6|
|82||F||Parotid||20 years||6.2||T4a||N0||M0||IVa||SC||FI||Surgery, RT||No||AWD||24|
|F||Parotid||10 years||EC||FI||Surgery||DM (lung)||DOD|
|62||F||SMG||18 years||N1||SDC||FI||Surgery, ND (pN+), RT||DM||DOD||12|
|72||M||Parotid||2 years||9.5||T4a||N1||M0||IVa||MC||FI||Surgery, ND (pN−), RT||No||AWD||58|
|66||M||Parotid||7 months||6||T3||N1||M0||III||SDC||FI||Surgery, ND (pN+), RT||RM||DOD||8|
|41||M||Parotid||15 years||6||T4a||N0||M0||IVa||AdNOS||FI||Surgery, ND (pN−)||No||DOC||9|
|51||M||Parotid||2 months||7||T3||N0||M0||III||AdNOS||FI||Surgery, ND (pN+), RT||Local||DOD||22|
|41||M||Parotid||16 years||3.5||T2||N1||M0||III||SDC||FI||Surgery, ND (pN+)||No||AWD||22|