Schwannoma with uncharacteristic rapid growth and high Ki-67 index: illustrative case

Abstract

BACKGROUND

Schwannomas typically show slow growth and a low risk of recurrence. The authors report a rare case of a 56-year-old woman with a spinal schwannoma that exhibited rapid recurrence and an atypically high Ki-67 index (8.9%), suggesting a more aggressive behavior than usually observed in benign schwannomas.

OBSERVATIONS

The patient initially presented with worsening low back and right sciatic nerve pain, resulting in difficulty walking independently. MRI revealed a dumbbell-shaped tumor compressing the cauda equina. A partial resection was performed to preserve nerve function, and pathology showed typical schwannoma features but with a Ki-67 index of 8.9%. The tumor rapidly regrew within 2 months, necessitating a second surgery and additional radiotherapy. Although MRI at 11 months after radiotherapy showed a tendency for tumor growth, subsequent MRI scans at 20 months after radiotherapy demonstrated a reduction in tumor size, indicating effective tumor control by the radiotherapy.

LESSONS

Among pathologically classified conventional schwannomas, those with a high Ki-67 index can have an atypical course that would not be considered benign. Adjuvant radiotherapy can be effective in controlling tumor growth in such atypical cases. Further research is recommended for managing such atypical cases.

https://thejns.org/doi/10.3171/CASE24529

Spinal schwannomas account for approximately 25%–30% of primary intradural spinal cord tumors in adults.^1–4 These tumors are typically slow-growing, and when complete removal is achieved through surgery, the risk of recurrence is low, estimated at around 4%–6%.³,^5–10 Recurrence, when it occurs, usually manifests several years after the initial surgery. Several risk factors for recurrence have been identified, including Sridhar classification, large tumor size, subtotal resection, and certain histopathological features.³,⁹ Among these, subtotal resection is a significant predictor of recurrence due to residual tumor cells. Additionally, the Ki-67 index, a marker of cellular proliferation, is generally low in conventional schwannomas, reflecting their benign nature.⁷,¹¹ However, cases exhibiting aggressive behavior, such as rapid growth and early recurrence, are exceptionally rare and challenge the traditional understanding of spinal schwannomas as benign tumors. High Ki-67 indices in schwannomas can indicate a higher proliferative potential and a propensity for rapid recurrence, which is not commonly expected in these tumors.

In this report, we present a rare case of a 56-year-old woman with a spinal schwannoma that demonstrated uncharacteristically rapid growth and a high Ki-67 index. The tumor recurred within 2 months after partial resection, necessitating a second surgery and adjuvant radiotherapy. Our case underscores the potential for aggressive behavior in schwannomas traditionally considered benign and highlights the need for close monitoring and consideration of additional therapeutic strategies in such atypical presentations.

Illustrative Case

A 56-year-old woman presented with low back pain and right sciatic nerve pain that worsened over time and resulted in difficulty walking independently, which required assistance with a cane, at 4 months after onset. The patient had no relevant past medical or familial history. Lumbar spine MRI performed by a physician at another center revealed a dumbbell-type tumor in the right nerve root of L3. She was then referred to our hospital.

On physical examination, the patient experienced numbness and dysesthesia from the right hip to the right outer lower limb, with the symptoms worsening during body movement. The manual muscle testing scores of the right iliopsoas, quadriceps, and hamstrings were 4, whereas for all other muscles, they were 5. Lumbar spine MRI revealed the presence of a dumbbell-type tumor in the right L3 nerve root, with a lesion in the spinal canal extending from the mid-inferior portion of the L2 vertebral body to the middle of the L3 vertebral body (Fig. 1). It also extended outside the right intervertebral foramen. The tumor size was 14 mm anteroposteriorly, 17 mm laterally, and 27 mm cephalocaudally (measured in each slice and shown at its maximum value), and the cauda equina was compressed to the left side by the tumor. The peripheral and solid components of the tumor demonstrated slightly lower signal intensity than the spinal cord on T1-weighted imaging (T1WI) and slightly higher signal intensity on T2-weighted imaging (T2WI). Prominent high-signal areas, indicative of necrosis or cystic degeneration, were observed in multiple locations within the lesion on T2WI. On gadolinium-enhanced T1WI, clear enhancement was noted in the peripheral and solid components. Bone invasion was not observed. For the preoperative diagnosis, a schwannoma was suspected.

FIG. 1.

Initial preoperative T1-weighted (A); T2-weighted (B); sagittal, coronal, and axial (tumor head, middle, and tail) Gd-enhanced T1-weighted (C–G) MR; and CT (H) images of the spinal cord. The tumor demonstrated slightly lower signal intensity than the spinal cord in the peripheral and solid areas on T1WI and slightly higher signal intensity on T2WI. Prominent high-signal areas were observed within the tumor, suggesting necrosis or cystic degeneration. On contrast-enhanced T1WI, the lesion within the spinal canal exhibited a ring-enhancing pattern, while the solid portion extending toward the caudal neural foramen showed relatively uniform enhancement, along with areas of cystic degeneration. No bone invasion was observed, and the neural foramen was expanded on the side of the tumor extension compared to the opposite side.

To improve the patient’s symptoms, tumor resection was performed via the posterior approach by using transcranial motor evoked potential (MEP)/somatosensory evoked potential and trigger MEP for intraoperative neuromonitoring. The intraspinal portion of the tumor was completely resected, and the portion outside the L3–4 right intervertebral foramen was resected as much as possible without resecting the facet and pars interarticularis (Fig. 2). Given that the trigger MEP showed that the right L3 motor branch was fused to the tumor within the intervertebral foramen and was difficult to separate, the tumor was removed as much as possible via an approach from outside the foramen, and the procedure was terminated when the MEP amplitude was reduced. There was no obvious postoperative worsening of the paralysis, and the pain disappeared. Histopathology showed that the tumor was composed of spindle-shaped cell proliferation with a coarse-to-dense cell density, such as Antoni A and B, and nuclear palisading (Fig. 3). The results of the immunohistochemical analysis were as follows: S100 (+), progesterone receptor (PgR) (−), epithelial membrane antigen (EMA) (−), somatostatin receptor 2 (−), Lys27 (+), and methylthioadenosine phosphorylase (MTAP) (+). The patient was then diagnosed with schwannoma, but the Ki-67 index was 8.9% in some hotspots.

FIG. 2.

Intraoperative images before (A) and after (B) tumor resection during the first operation. In both images, the right side corresponds to the cranial direction and the left side corresponds to the caudal direction. Before resection, the tumor compressed the cauda equina. After resection, the intradural tumor was completely removed while preserving the right L3–4 facet joint (*). The tumor extending into the intervertebral foramen was also resected as much as possible under neuromonitoring via an approach from outside the foramen (arrow).

FIG. 3.

Histopathology of the tumor during the first assessment (A–C) and during recurrence (D–F). The pathology of a conventional schwannoma is shown (A, B, D, and E), but the high Ki-67 index is atypical. H&E (A and D), MTAP (B and E), and Ki-67 (C and F) staining, original magnification ×200.

The patient was discharged without symptoms on the 8th postoperative day. However, the pain gradually reappeared from the right hip to the lateral front of the femoral region, and MRI showed a tumor recurrence in the same lesion at 2 months postoperatively (Fig. 4). At 7 months postoperatively, the tumor had increased to almost the same size as before the initial surgery (15 mm anteroposteriorly, 19 mm laterally, 24 mm cephalocaudally), and the degree of nerve compression was comparable. The MRI signal characteristics and contrast enhancement of the lesion were also similar to those before the initial surgery. Therefore, a second surgery was performed under intraoperative nerve monitoring as in the first surgery.

FIG. 4.

Sagittal (A, D, G, J, M, and P), axial (B, E, H, K, N, and Q), and coronal (F, I, L, O, and R) Gd-enhanced T1-weighted and axial T2-weighted (C) MR images of the spine obtained over time following the initial surgery. A–C: Five days after the initial surgery. The intradural portion of the tumor within the spinal canal was completely resected. The right side of the dura mater shifted toward the center of the spinal canal, and the contrast-enhanced lesion around it and those in the epidural space were interpreted, in conjunction with the intraoperative findings, as reflecting postoperative inflammatory changes. D–F: Two months postoperatively. Progression of the residual epidural tumor, centered at the right L3–4 foraminal level, was evident, with recurrence extending into the dural sac. G–I:Two days after the second surgery. Because of the spinal fixation, metal artifacts are present, making evaluation of the surgical site more challenging. Nevertheless, decompression of the dural sac appears to have been achieved. J–L: Three months after radiotherapy. The tumor shows a tendency for progression. M–O: Eleven months after radiotherapy. Continued tumor growth is observed, with an increase in the size of the lesion extending into the dural sac, suggesting ongoing progression despite radiotherapy. P–R: Twenty months following radiation therapy. Tumor shrinkage is evident compared with previous imaging, indicating a favorable therapeutic response following the transient increase observed after radiation therapy.

In the second surgery, the right L3–4 facet and pars interarticularis were resected. Although the surgical difficulty was high due to the strong adhesion of the granulation tissue, the tumor in the spinal canal was completely resected, and the tumor beyond the intervertebral foramen was resected gross totally under the capsule. Finally, to prevent joint instability, L3–4 TLIF was performed. Intraoperative neuromonitoring showed mild amplitude reduction in the quadriceps and hamstrings, but there was no obvious postoperative muscle weakness, and the numbness and dysesthesia were better than those experienced preoperatively. The results of the immunohistochemical stains were as follows: S100 (+), SMA (−), SOX10 (+), vimentin (+), desmin (−), CD34 (−), CD56 (+), CD57 (−), and P53 (−). The second diagnosis was schwannoma but with a partial Ki-67 index of 17%.

The tumor recurred after a short period and was considered to have a very high proliferative potential; therefore, radiotherapy was conducted to prevent recurrence. The patient received intensity-modulated radiotherapy at a dose of 50 Gy divided into 25 fractions. Three months after radiotherapy, the tumor had regrown, and by 11 months, it had increased further in size. However, at 20 months after radiotherapy, the tumor size had decreased compared with the 11-month images. Throughout this period, the patient remained asymptomatic with no neurological deficits.

Informed Consent

The necessary informed consent was obtained in this study.

Discussion

Observations

Spinal schwannoma is defined in the WHO classification of tumors, 5th edition as a benign nerve sheath tumor composed of differentiated neoplastic Schwann cells.¹² The incidence of spinal schwannomas is 0.3–0.4 cases per 100,000 persons per year.⁴ They are usually diagnosed during the 4th and 5th decades of life. Most spinal schwannomas are solitary and sporadic (95%); however, they are reported to be associated with neurofibromatosis type 2 (NF2). In the present case, there were no intracranial or other sites of neoplastic lesions, no cataracts, and no family history of NF2; therefore, the patient’s tumor was not an NF2 tumor.

The differential diagnoses of schwannoma include meningioma, neurofibroma, perineurioma, solitary fibrous tumor/hemangiopericytoma, malignant peripheral nerve sheath tumor (MPNST), and metastatic tumor. Although these tumors are often detected at preoperative stages by their location and characteristic findings on CT and MRI, it is often difficult to distinguish them from each other. Given that they are often detected owing to the appearance of the symptoms, the tumors are surgically removed, and a definitive diagnosis is made by histopathological examination. In the present case, the initial surgical specimen comprised spindle cells forming a tumor mass with varying densities, and findings corresponding to Antoni A and B were observed. Immunohistochemistry results were as follows: S100 (+), PgR (−), EMA (−), and negating meningioma. MTAP, which is often negative in sarcomas, was retained. Lys27 is a lysine residue of K27me3, but it becomes negative because of mutations. More than half of MPNSTs are negative, which is useful as an auxiliary diagnosis; however, in the present case, Lys27 was positive. In addition, immunohistochemical analysis performed at the time of reoperation showed the following: SMA (−), desmin (−), negating myoma, and sarcoma. The epithelial marker CD34 was negative, and S100 (+), SOX10 (+), and vimentin (+) supported the diagnosis of schwannoma.

In summary, the positivity for S100, SOX10, and vimentin strongly supported the diagnosis of schwannoma, whereas the negativity for other markers (PgR, EMA, SMA, desmin, and CD34) ruled out other possibilities (meningioma, myxoma, or sarcoma). Moreover, retention of MTAP and positivity for K27me3 were also reasons to exclude MPNST and sarcoma. The diagnostic criteria for MPNST remain unclear because no tumor markers or characteristic genetic mutations have been found. The gene mutations in NF1, CDKN2A, TP53, and PRC2 are reportedly involved in NF1-associated MPNSTs, but this is unclear in NF1-unassociated cases.¹³ From the abovementioned findings, these tumors are diagnosed as schwannomas.

In the present case, the patient was diagnosed with a conventional schwannoma. However, a characteristic feature of the present case is that the Ki-67 index was high at 8.9%, which is different from that of a typical schwannoma. A Ki-67 index > 5% suggests a malignant subtype.⁷ The term “malignant schwannoma” is often used, but its definition is not settled. According to the WHO classification, there are several subtypes of schwannomas, each with specific histological features. Cellular schwannomas are characterized by dense tissue images, with the Antoni A tissue being the main body and the density having a minimal difference. Melanotic schwannomas contain melanin pigments and are histologically clearly different. A plexiform schwannoma is often seen in the skin and subcutaneous tissue and exhibits a multinodular and creeping growth pattern. Epithelioid schwannomas show a multilobulated growth of epithelioid cells arranged singly or in nests within a variably myxoid or hyalinized stroma. Loss of SMARCB1 (INI1) expression is observed in approximately 40% of epithelioid schwannomas.¹⁴ Microcystic/reticular schwannomas are most commonly seen in the gastrointestinal tract. Microscopically, the tumor is characterized by a microcyst-rich network of interconnected benign spindle cells with an eosinophilic cytoplasm, associated with a myxoid, fibrillary, and/or hyalinized collagenous stroma. Our case did not fit any of these subtypes; hence, it was diagnosed as a conventional schwannoma. However, rapidly recurring schwannomas need to be classified separately from conventional schwannomas.

The risk factor for recurrence after undergoing surgical procedures includes subtotal resection.³,⁷Our patient had responded to trigger MEP during the initial surgery, which was terminated with partial resection to preserve the motor nerve, and the Ki-67 index was high at 8.9%, which placed her at a high risk for regrowth. As described above, the postoperative recurrence rate of conventional schwannomas is estimated to be 4%–6%, but Sohn et al. reported a regrowth of residual tumor after surgery in 29.6% of spinal schwannomas, with a mean Ki-67 index of 6.3%.¹¹ Spinal schwannomas are usually slow-growing, and the median recurrence time postoperatively ranged from 18.8 months to 27.5 months, as reported in the previous literature.³,⁹,¹⁵ According to Sohn et al., regenerative tumors grow at an average rate of 1.0 ± 4.4 mm/year.¹¹ However, the most aggressive tumor growth rate was 22.0 mm/year, and the Ki-67 labeling index was 17.1% in that report. In our case, the maximum diameter increased to 24 mm in 5 months and at a much faster rate. Recurrences on the order of months are often malignant and often NF2-related or MPNST. In a previous study on conventional schwannoma subtypes, elevated S100 and CD57 expression, along with a higher Ki-67 index, was linked to recurrence, whereas an increased P53 expression was correlated with reduced overall survival.⁷,¹¹ In the present case, the initial preoperative CT showed scalloping near the L3–4 right intervertebral foramen, suggesting that the tumor had originally grown slowly over a long course. This may indicate that the patient is undergoing a malignant transformation.

Complete resection is essential to prevent recurrence; however, when there is a risk of nerve damage, a strategy of maximum possible resection is adopted. It has been reported that radiotherapy can be used to prevent regrowth of any residual tumor.¹⁶ However, there have been no large studies, and the effectiveness of this treatment remains unclear. In the present case, although MRI at 11 months after radiotherapy indicated a potential increase in tumor size, subsequent imaging showed a reduction in tumor size. This suggests a delayed response to radiotherapy, which can occur due to the time required for radiation-induced tumor cell death and resorption. The initial apparent increase in size at 11 months might have been due to radiation-induced inflammation or pseudoprogression, a phenomenon where the tumor appears larger on imaging before eventual shrinkage.

Our case highlights that, in schwannomas with high Ki-67 indices, adjuvant radiotherapy can effectively control tumor growth, even if early postradiotherapy imaging suggests progression. Previous studies have shown mixed results regarding the efficacy of radiotherapy for recurrent or residual spinal schwannomas. However, our case supports the use of radiotherapy as a valuable adjunct treatment in managing schwannomas with high proliferative potential and aggressive behavior. The current classification methods categorize our case as a schwannoma, but we believe that, since the recurrence happened in a very short period, it cannot be considered a benign process. Thus, the present case should be classified separately from conventional schwannomas. Further studies are needed to establish guidelines on the timing of imaging follow-up and the interpretation of postradiotherapy changes in such cases.

Lessons

The tumor in the present case was diagnosed as a conventional schwannoma; however, the notable features were the elevated Ki-67 index at 8.9%, indicating a higher proliferative activity, and the unique situation with a rapid recurrence, which is uncommon in conventional schwannomas. Our case demonstrated that even in schwannomas, which are considered benign, a high Ki-67 index can lead to rapid recurrence; therefore, close follow-up is necessary for patients with this tumor. Additionally, although radiotherapy can help suppress tumor growth at the time of recurrence, it might require a prolonged period before any significant reduction in tumor size is observed. Given that there are no clear guidelines for managing such cases, especially those with atypical features, further larger studies are needed to determine the optimal treatment strategy.

Disclosures

Dr. Akazawa reported personal fees from Canon Medical Systems Corp. outside the submitted work.

Author Contributions

Conception and design: Umebayashi, Nishii, Hisaoka, Hashimoto. Acquisition of data: Umebayashi, Nishii, Kamata, Nagai, Shishido-Hara. Analysis and interpretation of data: Nishii, Shishido-Hara, Akazawa. Drafting the article: Umebayashi, Nishii. Critically revising the article: Umebayashi, Akazawa. Reviewed submitted version of manuscript: Hashimoto. Approved the final version of the manuscript on behalf of all authors: Umebayashi. Study supervision: Hashimoto.

Supplemental Information

Previous Presentations

Portions of this work were presented at the 14th Annual Meeting of Asia Spine, Nagoya, Aichi, Japan, June 17, 2023.

Correspondence

Daisuke Umebayashi: Graduate School of Medical Science, Kyoto Prefectural University of Medicine, Kyoto, Japan. umebayad@koto.kpu-m.ac.jp.