Inflammatory myofibroblastic tumor (IMT) – previously known as “inflammatory pseudotumor”- is composed of proliferating myofibroblasts with an admixed inflammatory cell infiltrate consisting of plasma cells, lymphocytes, and/or eosinophils and neutrophils [1–3]. It was initially considered benign and non-neoplastic. Later on, follow-up data showed a significant rate of local recurrence (around 25%) and metastases (less than 2%) [1]. The identification of ALK gene rearrangement in approximately 50% of cases provided additional support that IMT was a distinct neoplasm [2, 3]. It is currently classified by the WHO as a true neoplasm of intermediate malignancy [1]

Uterine Inflammatory myofibroblastic Tumor is rare. The first two cases were reported by Gilks et al in 1987[4]. In 2017, Mandato et al analyzed the articles published on IMT between 1987 and 2017, using PubMed as a database and they were able to find 72 cases of Uterine IMT [5]. The frequency is probably higher because many cases might have been misdiagnosed as leiomyomas or leiomyosarcomas [6]

In the uterus, the macroscopic, microscopic and immunohistochemical features of IMT show some overlap with those of other relatively more common mesenchymal neoplasms, mainly leiomyoma and leiomyosarcoma [3, 6, 7]. The misdiagnosis would deny the patient appropriate follow up after surgical resection or therapy with tyrosine kinase inhibitors in cases of unresectable, recurrent [5, 8] or metastatic tumors [6].

Macroscopic Examination

Total Abdominal Hysterectomy with Bilateral Salpingo-oophorectomy was performed. Cervix and endometrium were normal. A solid tan gray submucosal myometrial nodule measuring 8.5x8x6cm, exhibiting a whorled and trabeculated cut surface was identified. No myxoid or cystic areas were seen. The ovaries and tubes were unremarkable

Microscopic Examination

The nodule and its interface with the adjacent normal myometrium were extensively sampled for microscopic examination (14 blocks). Microscopically, it consisted of a mesenchymal neoplasm, composed of spindle shaped cells arranged mostly in compact intersecting fascicles with absent to minimal amount of intervening intercellular stroma (fascicular growth pattern), intimately admixed and merging with areas where the stroma is more abundant and myxoid as demonstrated by Alcian Blue stain (myxoid growth pattern) (Fig 1). The compact and the myxoid areas were unevenly distributed. Except for the presence of scattered rare lymphocytes, the compact growth pattern was indistinguishable from what is seen in benign classical leiomyoma (Fig 2). In the myxoid areas, the cells had loose “tissue culture-like” appearance and the lymphocytes and plasma cells were slightly more abundant (Fig 3). The spindle cells in both areas appeared similar and displayed uniform oval nuclei, regular nuclear membrane, pale open chromatin and small amount of pale eosinophilic cytoplasm and occasional one or two inconspicuous nucleoli (minimal atypia). The interface between the nodule and the surrounding myometrium was irregular (geographic interface). There was no necrosis. Mitotic figures did not exceed 4 per 10 High Power Fields.

Figure 1: Intimate admixture of areas exhibiting a compact / fascicular growth pattern (open star) and areas of myxoid and loose growth pattern (black star) (H&E; original magnification; 25x)

Figure 2: Compact growth pattern. In the lower half of this field, given the absence of lymphocytes, myxoid change and atypia, the appearance is practically indistinguishable from that of classical leiomyoma. Transition to myxoid / loose area and rare scattered lymphocytes are seen in the upper half (H&E; original magnification; 100x)

Figure 3: Myxoid growth pattern. Abundant myxoid stroma, lower cellularity and slightly more abundant lymphocytes when compared with areas of compact fascicular growth pattern.  Cells have loose “tissue culture-like” appearance and minimal atypia (H&E; original magnification; 200x).

Immunohistochemistry

The following primary monoclonal antibodies were used: Smooth Muscle Actin (clone 1A4, prediluted; Dako), Desmin (clone D33, prediluted, Dako), H-Caldesmon, (clone h-CD, prediluted, Dako), CD10 (clone 56C6, prediluted, Dako), ALK (clone ALK1, prediluted, Dako) and p53 (clone DO-7, prediluted, Dako),

The cells expressed Smooth Muscle Actin (SMA) and Desmin (strong and diffuse expression). H-Caldesmon was expressed by rare scattered cells in the areas with fascicular growth. However, there was complete lack of expression in the areas with myxoid growth pattern. CD10 was negative. Granular cytoplasmic staining was seen with ALK-1 immunostain, mainly in the areas with a myxoid growth pattern (Fig 4). Weak nuclear expression of p53 was present in less than 10% of the cells.

Figure 4: ALK-1 immunohistochemical stain. Granular cytoplasmic staining (Hematoxylin counterstain; original magnification; 200x)

Fluorescent In Situ Hybridization (FISH)

Fluorescent In Situ Hybridization (FISH) to detect ALK gene rearrangements was performed using the Vysis LSI ALK Dual Color, Break Apart Rearrangement Probe (Abbott Molecular). This mix consists of two fluorophore-labeled DNA probes whose targets are on opposite sides, flanking the breakpoint of the ALK gene on the short arm of chromosome 2.

Following the manufacturer’s instructions, 3µm thick sections from Formalin Fixed Paraffin Embedded (FFPE) blocks of tissue were deparaffinized, dehydrated and washed. The sections were baked overnight at 60°C on a hot plate, immersed in Hemo-De for 15 min, treated with pretreatment solution (Abbott Molecular) at 80°C for 12 min, and protease buffer for 20 min at 37°C. After applying the probe onto the tissue sections, sealed slides were incubated in a humidified atmosphere on Hybrite (Abbott Molecular) at 73°C for 3 min, then incubated overnight at 37°C. The following day, slides were immersed in washing buffer and DAPI was applied for nuclear counterstaining. Signals were evaluated under a fluorescent microscope. A total of 200 nuclei were screened for the presence of adjacent (normal) or separate (abnormal) fluorescent signals. A total of 166 nuclei (81%) showed an abnormal patter. With the cut off being 15%), the sample was therefore reported as positive for ALK gene rearrangement.

A diagnosis of Inflammatory Myofibroblastic Tumor was rendered. The patient did not receive any additional treatment. She is not known to have presented for investigations of any other neoplasms 3 years after resection

Clinical features

Uterine IMTs present at a mean age of 40. They also seem to be associated with pregnancy [3, 6]. Presenting features include vaginal bleeding, abdominal/pelvic pain, abdominal distension. Nearly half of patients may be asymptomatic and the tumor is discovered incidentally [3, 5]. A constitutional syndrome consisting of fever, weight loss and malaise is seen in some, due to overproduction of IL-6 [9, 10] and laboratory evaluation may reveal microcytic anemia, raised ESR, thrombocytosis, and/or polyclonal hypergammaglobulinemia [11]. Interestingly, these systemic manifestations resolve following surgical excision, as previously reported in other locations [12] and tumor recurrence may be heralded by a return of clinical and laboratory abnormalities [2, 5]

Macroscopic appearance

Approximately 90% of the tumors arise from the corpus and 10% from the cervix, with a size ranging from 1 to 20 cm (mean of 6.8cm) [3, 5]. They appear as solid myometrial nodules, which may be submucosal (half of cases), intramural, or subserosal [3]. They often exhibit a tan gray fasciculated whorled cut surface and are indistinguishable from leiomyomas. Soft areas with myxoid consistency can be seen, and their extent is widely variable, accounting in some cases for about 90% of the tumor. Areas of cystic change, foci of hemorrhage, calcification and necrosis are identified in a minority of cases. [3, 5]. Their borders may range from being smooth to irregular or infiltrative [3].

Histology

Three basic histologic patterns have been described and are seen in various combinations within the same tumor [11]. The myxoid pattern is the most common. It is hypocellular and has a fasciitis or granulation tissue-like appearance, characterized by loosely arranged plump to spindle cells in an edematous or myxoid stroma and a prominent network of delicate blood vessels. The spindle cell pattern is compact. It is hypercellular and characterized by fascicular or storiform arrangement of spindle cells with elongated plump nuclei resembling those of smooth muscle cells set and within a collagenous stroma. The fibromatosis-like pattern is relatively hypocellular, characterized by elongated spindle cells in a background of hyalinized, dense collagen containing scattered lymphocytes, plasma cells and eosinophils. The spindle cells of IMT are mostly uniform and predominantly myofibroblastic in appearance, with pale eosinophilic cytoplasm, plump ovoid to tapering vesicular nuclei and one or two small nucleoli. Ganglion like cells (ovoid cells with eccentric nuclei, prominent nucleoli, and eosinophilic cytoplasm) are present in 15% to 50% of cases.

In the uterus, the differential diagnosis of IMT includes mainly smooth muscle neoplasms (leiomyomas and leiomyosarcomas) and endometrial stromal sarcomas (ESS). The fascicular areas of IMTs, consisting of spindle cells with elongated plump nuclei morphologically resemble those of smooth muscle cell neoplasms. Expression of actin and desmin further contribute to this similarity [3, 6, 13]. A helpful feature is the presence in IMT of at least focal myxoid areas often with a loose “tissue culture-like” appearance and interspersed lymphoplasmacytic infiltrate [6]. However, myxoid stroma can also be a feature of leiomyoma, leiomyosarcoma and endometrial stromal sarcoma [6, 7, 13]. Irregular infiltrative borders are seen in myxoid leiomyosarcoma, endometrial stromal sarcoma and IMT, whereas leiomyoma is well circumscribed [7, 13]. Leiomyosarcomas lack the inflammatory cell infiltrate and the tissue culture-like appearance and they exhibit moderate to severe atypia [3, 7, 13].

IMTs may show areas of uniform cells in diffuse sheets resembling endometrial stroma [3]. The infiltrative border and the possible expression of CD10 can lead to a misdiagnosis of endometrial stromal sarcoma (ESS) [3, 13]. However, IMT lacks the spiral arterioles commonly seen in ESS. Endometrial stromal sarcoma might rarely show areas of myxoid change but they lack the characteristic lymphoplasmacytic infiltrate of IMT and are negative for ALK [3, 13].

Immunohistochemistry

Consistent with its myofibroblastic differentiation, Uterine IMT show immunoreactivity for smooth muscle actin and desmin in 89.6% and 82.7% of cases respectively [5].H-Caldesmon is expressed in 38-42% of cases [5, 14, 15] CD10 in 30 to 90% of cases [6, 14–16] and Cytokeratins in 16-30% [5, 16]

Evaluation of p53 and p16 staining by immunohistochemistry can be useful in the distinction between uterine leiomyosarcoma and inflammatory myofibroblastic tumor as nearly half of leiomyosarcomas show abnormal p53 and p16 staining patterns whereas IMTs generally don’t [13, 17]. An abnormal p53 result is defined as strong nuclear staining in ≥80% tumor cells, or complete absence of staining. An abnormal p16 result is defined as strong and diffuse, nuclear and cytoplasmic positivity or complete absence of staining [13, 17].

Cytogenetics

ALK expression and gene rearrangement as demonstrated by immunohistochemistry (IHC) and Fluorescence In Situ Hybridization (FISH) respectively are diagnostic of IMT [3, 6, 14]. In uterine IMTs the rate of ALK positivity by immunohistochemistry ranges from 88% to 100% [3, 10, 14–16]. The smooth muscle like areas demonstrate diffuse strong ALK expression and harbor gene rearrangements by Fluorescent In Situ Hybridization (FISH), confirming that they are part of the neoplasm [3, 6, 13].

Approximately 50% of all IMTs and up to 80% of uterine IMTs show ALK rearrangements by FISH [2, 5, 16]. Multiple ALK fusion partner genes have been recognized including TPM3, SEC31, IGFBP5, TIMP3, THBS1, and DES [3, 6, 14]. Uterine IMTs are rich in ALK fusions with novel ALK partners IGFBP5 and THBS1 [3, 18]. Rearrangements of genes encoding other tyrosine kinases including ROS1, NTRK3, PDGFR-ß, and RET have been reported in many of the remaining extrauterine IMTs, although they have not been found in those arising in the uterus [3, 6]. Patterns of immunostaining have been correlated with specific gene fusions depending on the location of the fusion partner such as nuclear membrane staining with RANBP2-ALK fusion, diffuse cytoplasmic staining for TPM3, TPM4, CARS, ATIC and SEC31L1 fusion partners and granular cytoplasmic staining with the CLTC fusion partner. In uterine IMTs all cases have shown granular cytoplasmic staining pattern [2, 3, 19]. FISH results can be falsely negative when the fusion partners are on the same chromosome (cases harboring IGFBP5-ALK fusion, where both genes reside on chromosome 2) most likely because the spatial separation between the fluorescent signals in a break apart FISH assay is subtle and not appreciated [3, 18]

Prognosis

Inflammatory Myofibroblastic Tumor is considered of intermediate malignant potential [1]. Recurrence rate varies by anatomical site, from less than 2% for tumors confined to the lung [20, 21] to 25% for extrapulmonary lesions [2, 5, 11]. Recurrences are particularly common among multinodular intra-abdominal tumors, likely reflecting the difficulty of complete surgical resection [2, 5]. Metastasis is rare, occurring in 2-5% of cases [5]. The most common sites of metastasis are lung and brain, followed by liver and bone [5, 22].

The majority of gynecologic IMTs, even those treated by simple myomectomy or enucleation alone, do not recur [6]. Some metastasizing cases of uterine IMT were initially misdiagnosed as leiomyosarcomas [6, 7]

To date, morphologic parameters that predict behavior in a specific case of IMT are inconclusive. Certain features such as necrosis, large size (>7 cm), moderate to severe atypia, high mitotic activity (>10MF /10 HPFs) and infiltrative borders seem to be associated with an aggressive outcome [14, 15].

None of the previously mentioned criteria is sensitive nor specific enough in order to predict the behavior of a given tumor [22]. Necrosis is rare and although in some series it was only seen in aggressive cases [6, 15], it was also reported in up to 44% of clinically benign tumors [14, 22]. Many reported clinically benign cases measured more than 7cm (8 to 15cm) [4, 6, 10, 14, 18, 23]. Mild nuclear atypia can often be seen in biologically aggressive cases [6, 15, 24]. Mitotic count below 10 per 10HPF (1-8 MF/10HPFs) does not rule out the possibility of an aggressive outcome [6, 7, 14, 15, 24]

A predominance of myxoid pattern (90% or more) was reportedly associated with an aggressive behavior in one study [15]. However, this was not found in other series and another study found that the association between compact growth pattern and aggressive behavior approached statistical significance [14]. Ganglion like cells, the presence of which was suggested as indicative of aggressive behavior, are also seen in more than one half of clinically benign cases [2, 22].

However, if any of the previously mentioned worrisome features is present, it is recommended to report it so that the patient is followed closely. Immunoreactivity with anti p53 has an uncertain role in predicting the behavior of IMTs. It was detected in 25% of cases that have metastasized, compared with 81- 100% in classic, histologically atypical, and recurrent ones [22]. Unlike Anaplastic Large Cell Lymphoma (ALCL), in which ALK-positive tumors pursue a less aggressive clinical course, ALK negative IMT cases are associated with distant metastases [22].

Uterine Inflammatory Myofibroblastic tumor can closely mimic smooth muscle neoplasms, both grossly and microscopically. The distinction between these entities is crucial, given the difference in biology, prognosis and treatment.  A high index of suspicion is needed especially when a myxoid matrix is present and /or an admixed chronic inflammatory cell infiltrate is present. Immunohistochemical staining with ALK-1 antibody and Fluorescent in Situ Hybridization (FISH) are helpful with a reported sensitivity of 88%-100% and 80% respectively.

Conflicts of interest

The authors declare that the article was written in the absence of any financial, commercial, legal, or professional relationship that could be construed as a potential conflict of interest.

Acknowledgments

We would like to thank Laila F. Zahed PhD and Sara F. Korban BSc (Clinical Laboratories Department, Cytogenetics Section, Saint George Hospital University Medical Center) for the performance and interpretation of the Fluorescent In Situ Hybridization (FISH) test.

1. Coffin and J.A. Fletcher: “WHO Classification of Tumours of Soft Tissue and Bone.” IARC Press, Lyon. doi: 10.32074/1591-951X-213
2. C. Gleason and J.L. Hornick: “Inflammatory myofibroblastic tumours: where are we now?” J Clin Pathol. vol. 61, no. 4, pp. 428–437, 2008. https://doi.org/10.1136/jcp.2007.049387
3. S. Shukla and K. Mittal: “Inflammatory Myofibroblastic Tumor in Female Genital Tract.” Archives of Pathology & Laboratory Medicine. vol. 143, no. 1, pp. 122–129, 2019.  doi:10.5858/arpa.2017-0575-RA
4. B. Gilks, G.P. Taylor, and P.B. Clement: “Inflammatory Pseudotumor of the Uterus:” International Journal of Gynecological Pathology. vol. 6, no. 3, pp. 275–286, 1987.
5. D. Mandato, R. Valli, V. Mastrofilippo, A. Bisagni, L. Aguzzoli, and G.B. La Sala: “Uterine inflammatory myofibroblastic tumor: more common than expected.” Medicine. vol. 96, no. 48, pp. e8974, 2017. doi: 10.1097/MD.0000000000008974
6. L. Pickett, A. Chou, J.A. Andrici, A. Clarkson, L. Sioson, A. Sheen, J. Reagh, F. Najdawi, Y. Kim, D. Riley, J. Maidens, D. Nevell, K. McIlroy, S. Valmadre, G. Gard, R. Hogg, J. Turchini, G. Robertson, M. Friedlander, and A.J. Gill: “Inflammatory Myofibroblastic Tumors of the Female Genital Tract Are Under-recognized: A Low Threshold for ALK Immunohistochemistry Is Required.” The American Journal of Surgical Pathology. vol. 41, no. 10, pp. 1433–1442, 2017. doi: 10.1097/PAS.0000000000000909
7. Parra-Herran, J.K. Schoolmeester, L. Yuan, P. Dal Cin, C.D.M. Fletcher, B.J. Quade, and M.R. Nucci: “Myxoid Leiomyosarcoma of the Uterus: A Clinicopathologic Analysis of 30 Cases and Review of the Literature With Reappraisal of Its Distinction From Other Uterine Myxoid Mesenchymal Neoplasms.” American Journal of Surgical Pathology. vol. 40, no. 3, pp. 285–301, 2016. doi: 10.1097/PAS.0000000000000593
8. E. Butrynski, D.R. D’Adamo, J.L. Hornick, P. Dal Cin, C.R. Antonescu, S.C. Jhanwar, M. Ladanyi, M. Capelletti, S.J. Rodig, N. Ramaiya, E.L. Kwak, J.W. Clark, K.D. Wilner, J.G. Christensen, P.A. Jänne, R.G. Maki, G.D. Demetri, and G.I. Shapiro: “Crizotinib in ALK -Rearranged Inflammatory Myofibroblastic Tumor.” N Engl J Med. vol. 363, no. 18, pp. 1727–1733, 2010. doi: 10.1056/NEJMoa1007056
9. Azuno, K. Yaga, Y. Suehiro, S. Ariyama, and A. Oga: “Inflammatory myoblastic tumor of the uterus and interleukin-6.” American Journal of Obstetrics and Gynecology. vol. 189, no. 3, pp. 890–891, 2003. https://doi.org/10.1067/S0002-9378(03)00208-4.
10. T. Rabban, C.J. Zaloudek, K.M. Shekitka, and F.A. Tavassoli: “Inflammatory Myofibroblastic Tumor of the Uterus.” Am J Surg Pathol. vol. 29, no. 10, pp. 8, 2005. doi: 10.1097/01.pas.0000172189.02424.91
11. M. Coffin, J. Watterson, J.R. Priest, and L.P. Dehner: “Extrapulmonary inflammatory myofibroblastic tumor (inflammatory pseudotumor). A clinicopathologic and immunohistochemical study of 84 cases.” Am J Surg Pathol. vol. 19, no. 8, pp. 859–872, 1995. 10.1097/00000478-199508000-00001
12. J. Chouairy, E.A. Bechara, S.J. Gebran, and R.H. Ghabril: “Inflammatory myofibroblastic tumor of the mesentery associated with high fever and positive Widal test.” Saudi Med J. vol. 29, no. 12, pp. 1819–1823, 2008.
13. Busca and C. Parra-Herran: “Myxoid Mesenchymal Tumors of the Uterus: An Update on Classification, Definitions, and Differential Diagnosis.” Adv Anat Pathol. vol. 24, no. 6, pp. 8, 2017. doi: 10.1097/PAP.0000000000000164
14. A. Bennett, V. Nardi, M. Rouzbahman, V. Morales-Oyarvide, G.P. Nielsen, and E. Oliva: “Inflammatory myofibroblastic tumor of the uterus: a clinicopathological, immunohistochemical, and molecular analysis of 13 cases highlighting their broad morphologic spectrum.” Mod Pathol. vol. 30, no. 10, pp. 1489–1503, 2017. DOI:10.1038/modpathol.2017.69
15. Parra-Herran, C.M. Quick, B. E. Howitt, P.D. Cin, B.J. Quade and M.R. Nucci : “Inflammatory Myofibroblastic Tumor of the Uterus: Clinical and Pathologic Review of 10 Cases Including a Subset With Aggressive Clinical Course..” Am J Surg Pathol. vol. 39, no. 2, pp. 12, 2015. doi: 10.1097/PAS.0000000000000330
16. E. Fuehrer, G.L. Keeney, R.P. Ketterling, R.A. Knudson, and D.A. Bell: “ALK-1 protein expression and ALK gene rearrangements aid in the diagnosis of inflammatory myofibroblastic tumors of the female genital tract.” Arch Pathol Lab Med. vol. 136, no. 6, pp. 623–626, 2012. http://ezsecureaccess.balamand.edu.lb/login?url=https://www.proquest.com/scholarly-journals/alk-1-protein-expression-gene-rearrangements-aid/docview/1362248394/se-2
17. -M. Schaefer, J.L. Hornick, L.M. Sholl, B.J. Quade, M.R. Nucci, and C. Parra-Herran: “Abnormal p53 and p16 staining patterns distinguish uterine leiomyosarcoma from inflammatory myofibroblastic tumour.” Histopathology. vol. 70, no. 7, pp. 1138–1146, 2017.doi: 10.1111/his.13176
18. D. Haimes, C.J.R. Stewart, B.A. Kudlow, B.P. Culver, B. Meng, E. Koay, A. Whitehouse, N. Cope, J.-C. Lee, T. Ng, W.G. McCluggage, and C.-H. Lee: “Uterine Inflammatory Myofibroblastic Tumors Frequently Harbor ALK Fusions With IGFBP5 and THBS1:” The American Journal of Surgical Pathology. vol. 41, no. 6, pp. 773–780, 2017. doi: 10.1097/PAS.0000000000000801
19. R. Cook, L.P. Dehner, M.H. Collins, Z. Ma, S.W. Morris, C.M. Coffin, and D.A. Hill: “Anaplastic Lymphoma Kinase (ALK) Expression in the Inflammatory Myofibroblastic Tumor: A Comparative Immunohistochemical Study.” The American Journal of Surgical Pathology. vol. 25, no. 11, pp. 1364–1371, 2001.
20. J. Cerfolio, M.S. Allen, A.G. Nascimento, C. Deschamps, V.F. Trastek, D.L. Miller, and P.C. Pairolero: “Inflammatory pseudotumors of the lung.” The Annals of Thoracic Surgery. vol. 67, no. 4, pp. 933–936, 1999. doi:10.1016/S0003-4975(99)00155-1
21. S. Janik, J.P. Janik, M.A. Lovell, R.J. Hendrickson, D.D. Bensard, and B.S. Greffe: “Recurrent inflammatory pseudotumors in children.” Journal of Pediatric Surgery. vol. 38, no. 10, pp. 1491–1495, 2003. doi:10.1016/s0022-3468(03)00501-3
22. M. Coffin, J.L. Hornick, and C.D.M. Fletcher: “Inflammatory Myofibroblastic Tumor.” Am J Surg Pathol. vol. 31, no. 4, pp. 12, 2007.doi: 10.1097/01.pas.0000213393.57322.c7
23. Shintaku and A. Fukushima: “Inflammatory myofibroblastic tumor of the uterus with prominent myxoid change.” Pathol Int. vol. 56, no. 10, pp. 625–628, 2006. doi: 10.1111/j.1440-1827.2006.02018.x
24. Fraggetta, C. Doglioni, P. Scollo, L. Pecciarini, M. Ippolito, P. Amico, G. Pelosi, and M. Ponzoni: “Uterine Inflammatory Myofibroblastic Tumor in a 10-Year-Old Girl Presenting As Polypoid Mass.” JCO. vol. 33, no. 2, pp. e7–e10, 2015. doi:10.1200/JCO.2013.48.830