Twenty-Month Regression Following Concurrent Conventional Whole-Brain Irradiation and Chemoimmunotherapy for ≥3.8 cm Cerebellar Metastasis From Small Cell Lung Cancer

Standard whole-brain radiotherapy (WBRT) alone for large brain metastases (BMs) from small cell lung cancer (SCLC) has limited efficacy and durability, and stereotactic radiosurgery (SRS) alone for symptomatic posterior fossa BMs >3 cm with satellite lesions is challenging. Herein, we describe the case of a 73-year-old female presenting with treatment-naïve SCLC and 15 symptomatic multiple BMs, including a ≥3.8-cm cerebellar mass (≥17.7 cm3) and two adjacent lesions; otherwise, the SCLC was confined to the thorax. The patient was initially treated concurrently with conventional WBRT (30 Gy in 10 fractions) without boost and chemoimmunotherapy (CIT) consisting of carboplatin, etoposide, and atezolizumab. Atezolizumab was excluded during irradiation. Five months after WBRT, the large cerebellar lesion had remarkably regressed, and the smaller lesions (≤17 mm) showed complete responses (CRs) without local progression at 20 months. However, six and 16 months after WBRT, the thoracic lesions had progressed, and although amrubicin was administered, four new BMs, including pons involvement, had developed, respectively. Despite the CRs of the four BMs following SRS (49.6 Gy in eight fractions) and the sustained regression of the thoracic lesions, meningeal dissemination and multiple new BMs were evident 3.5 months post-SRS. The small remnant of the large BM and/or newly developed BMs abutting the cerebrospinal fluid (CSF) space could have led to CSF dissemination, the presumed cause of the patient’s death. Taken together, concurrent chemo-WBRT and subsequent CIT can provide excellent and durable tumor responses for SCLC BMs, but may not be fully sufficient for BMs ≥3.8 cm. Therefore, in cases with large lesions, focal dose escalation of the large lesions, consolidative thoracic radiotherapy, and dose de-escalation in the macroscopically unaffected brain region may prevent or attenuate CSF dissemination, new BM development, and adverse effects and thus should be considered.


Introduction
Small cell lung cancer (SCLC) is an aggressive neuroendocrine malignancy with a high predisposition to developing brain metastasis (BM) [1,2]. The brain is a sanctuary site for metastatic cells owing to poor penetration of chemotherapeutic drugs across the blood-brain and blood-tumor barriers. Hence, external beam radiotherapy (EBRT) is an essential treatment option for BMs from SCLCs.
Owing to the one-time application and the adverse effects of WBRT even with reduced prophylactic doses and driven by advances in systemic therapy, treatment strategies for small SCLC-BMs using upfront SRS with preservation of WBRT are attracting attention [5,8,9]. However, SRS for large (>3 cm) symptomatic posterior fossa BMs with satellite lesions inevitably increases dose spillage in the surrounding brain tissue and the risk of complications (e.g., obstructive hydrocephalus and tonsillar herniation due to radiationinduced edemata of the surrounding brain tissue and/or the tumor itself). Thus, the effective and safe 1, 2 3, 4 4 4 2 performance of SRS in such scenarios remains challenging [10]. Furthermore, given the frequency and severity of microscopic brain invasion by SCLC-BMs [2,10,11], ensuring safe long-term local control of BMs >3.5 cm using only SRS in less than five fractions remains an unsolved conundrum [10].
The response of BMs to CIT without EBRT remains unclear and is still under investigation [12]. For symptomatic and/or large SCLC-BMs, EBRT or other local therapy is generally performed before systemic therapy, whereas the efficacy and safety of concurrent WBRT and CIT are unresolved [1,2,8,13]. Lastly, for symptomatic large BMs not amenable to surgical removal, steroids must be administered to ameliorate and stabilize neurological symptoms before administration of immune checkpoint inhibitors (ICIs); however, steroids may suppress CD8+ cytotoxic T-cell activity and hamper the efficacy of the ICIs [14].
Herein, we describe the case of a 73-year-old female who presented with treatment-naïve ES-SCLC and multiple symptomatic BMs, including a >3.8-cm cerebellar lesion and two adjacent lesions. Concurrent conventional WBRT without local boost and CIT resulted in remarkable regression of the BMs without local progression for 20 months. The SCLC lesions in the thorax became refractory within six months after WBRT but did not progress further; however, the patient died, most likely due to cerebrospinal fluid (CSF) dissemination. We discuss possible measures for improving BM control, preventing CSF dissemination, and attenuating treatment-related neurotoxicity in the CIT era.
This report was part of the clinical study approved by the Clinical Research Review Board of Kainan Hospital Aichi Prefectural Welfare Federation of Agricultural Cooperatives (20220727-1).

Case Presentation
A 73-year-old female presented with unsteadiness, headaches, and difficulty walking. The patient was a current smoker, and the past medical history was unremarkable. No abnormalities were noted on screening performed two years earlier. The Karnofsky performance scale (KPS) score was 60 at consultation, and the patient was directly admitted to the hospital.   CE: contrast-enhanced; WIs: weighted images; WBRT: whole-brain radiotherapy Thoracic CE computed tomography revealed moderate centrilobular emphysema with a perihilar mass lesion and multiple swollen mediastinal lymph nodes ( Figure 3).

FIGURE 3: Chest computed tomography images at the initial diagnosis
The images show axial CE computed tomography (CT) images (A-D).
(A-D) The 2-cm well-demarcated mass lesion in the right middle-inferior perihilar region (arrow in A) and multiple lymph node swelling in the ipsilateral hilar and peribronchial (arrows in B), subcarinal (arrow in C), and subaortic (arrow in D) regions.

CE: contrast-enhanced
The serum tumor marker elevations included pro-gastrin-releasing peptide (ProGRP) of 3121 pg/mL (normal limit, ≤80 pg/mL), neuron-specific enolase (NSE) of 19.4 ng/ml (≤15.0), and carcinoembryonic antigen of 11.3 ng/ml (≤4.1). The symptoms resolved after the administration of steroids and glycerol. Small cell lung cancer was pathologically diagnosed via a bronchoscopic biopsy. The clinical stage was IV B (cT1b N3 M1c) based on the tumor, node, and metastasis (TNM) classification according to the eighth edition of the Union for International Cancer Control (UICC). For the ES-SCLC, brain radiotherapy and CIT were concurrently performed [2,8,11,13]. The anti-cancer treatments are summarized in Figure 4; changes in ProGRP levels over time are also shown. Thirteen days after the initial MRI scan, the neurological symptoms had stabilized. At this time, WBRT (30 Gy in 10 fractions) with two-dimensional irradiation using two opposed fields was initiated; the prescription dose was assigned to the isocenter itself. The treatment platform was a multileaf collimator Agility® (Elekta AB, Stockholm, Sweden) mounted in a linac Infinity® (Elekta AB, Stockholm, Sweden) with a 6 megavoltage (MV) X-ray beam. The planning system was Monaco® (Elekta AB, Stockholm, Sweden), and the dose calculation algorithm was a collapsed cone with a calculation grid size of 2 mm. The treatment planning was performed by the predecessor. The dose distribution and planning parameters for the large cerebellar lesion are shown in Figure 5. Etoposide plus carboplatin (EC) was administered one day after WBRT initiation, and atezolizumab was administered seven days after WBRT completion ( Figure 4). These treatments were tolerated and completed by the patient without significant adverse events.
Following chemo-WBRT, the ProGRP level decreased to 61.8 pg/mL, and the thoracic lesions showed a partial response at the beginning of the fourth course of EC and the third course of atezolizumab. An MRI at 4.9 months after WBRT initiation revealed remarkable regression of the large cerebellar lesion, leaving only a small remnant, and complete responses (CRs) for the other 14 brain lesions (all ≤17 mm) (Figures 1, 2). However, during the eighth course of atezolizumab (5.8 months following WBRT), the disease was deemed refractory owing to the elevation of the ProGRP level and enlargement of the peribronchial lymph node metastases ( Figure 4). Therefore, atezolizumab in the maintenance phase was replaced by amrubicin (AMR), which resulted in favorable tumor marker and imaging responses, with the nadir ProGRP response at 7.2 months (Figure 4). AMR was administered 11 times until progression.
At 15.9 months, the ProGRP level increased without progression of the thoracic lesions. An MRI performed at this time revealed four new BMs, including the pons involvement and those contacting the CSF space, but no local progression of the initial 15 BMs (Figures 6-7).  CE: contrast-enhanced; WIs: weighted images; mo: months; WBRT: whole-brain radiotherapy; DVHs: dosevolume histograms; GTV: gross tumor volume; GTV + 2 mm: the reference volume generated by adding an isotropic 2-mm margin to a GTV Amrubicin monotherapy was switched to irinotecan (CPT-11) monotherapy at 16.4 months, and the four new BMs were treated via SRS ( Figure 7); owing to brainstem involvement and prior WBRT, the prescription dose was 49.6 Gy in eight fractions to the gross tumor volume (GTV) margin with 67%-78% isodose coverage to ensure both efficacy and safety. Stereotactic radiosurgery was implemented with volumetric modulated arcs (VMA) using the Agility® on the Infinity®, in which the four lesions were simultaneously irradiated with a flattening filter-free mode of a 6 MV X-ray beam via a single isocenter [15]. The arc arrangement consists of one coplanar arc and two non-coplanar arcs with each arc length of 180º, which are allocated at 60º intervals to divide the cranial hemisphere evenly. The collimator angles for each arc are separately set to be 90º, 45º, and 135º. The dedicated software MIM Maestro TM (MIM Software, Cleveland, OH, USA) was used for image co-registration, fusion, and contouring [10]. Monaco® was used to optimize the VMA plan ( Table  1).   The patient was treated for febrile neutropenia 19 days after the initiation of CPT-11 monotherapy and SRS. An MRI at 2.1 months after CPT-11/SRS initiation showed CRs for the four new lesions, with sustained regression at 3.5 months (Figure 7). However, an MRI at 20 months after WBRT (3.5 months after SRS) revealed limited meningeal dissemination along with multiple new lesions (Figure 2), although none of the initial 15 BMs had progressed locally (Figures 1, 2, 6). Cross-temporal comparison of the brain morphology on serial MRI scans revealed obvious progression of ventricular dilatation and widening of the cortical sulci, both of which were mainly attributed to a decrease in brain volume ( Figure 8). WIs: weighted images; CE: contrast-enhanced; WBRT: whole-brain radiotherapy Owing to the presence of CSF dissemination, resistance to CPT-11 monotherapy, and declining performance status, the patient was transitioned to palliative care. One month later, oral intake became difficult, and the ProGRP level increased precipitously (NSE at 21.2 months, 38.0 ng/mL), and the patient died 21.5 months after the initiation of concurrent chemo-WBRT (Figure 4).

Discussion
The present case describes the challenging treatment of at least 15 synchronous SCLC-BMs, including a symptomatic ≥3.8-cm cerebellar lesion, in a current smoker in her early 70s. The cerebellar lesion was likely >4 cm (>20 cm 3 ) at the start of WBRT given the 13-day interval between its initial detection and WBRT initiation, although the extracranial tumor burden was asymptomatic and confined to the thorax [3,11,16].
As proposed in 2018, the brain metastases from SCLC (BMS) score serves as the prognostic score for WBRTtreated SCLC-BMs [17]. The BMS score in the present case was "group II" (two points), and, assuming the extracranial lesion was stable, the estimated median survival time was 6.6 months. According to the National Cancer Database Survey from 2010 to 2014, the median survival times for older patients (≥75 years) with SCLC-BMs who received WBRT alone, chemotherapy with WBRT, and chemotherapy without WBRT were 1.9, 5.6, and 6.4 months, respectively; the difference between WBRT with/without chemotherapy was not significant [16]. Thus, in the pre-CIT era, the outcomes of older patients with SCLC-BMs, especially those with low KPS scores, were usually dismal, irrespective of treatment selection.
Against this background, the excellent and durable responses of the initial BMs in the present case likely reflect the significant enhancement of conventional WBRT by concurrent EC treatment and subsequent EC treatment plus an ICI. Achieving similar efficacy and safety when using SRS (less than five fractions) for large BMs is not as straightforward. Among the 36 pre-CIT era cases of ES-SCLC reported by Chen et al., in which the brain was the sole distant metastatic site, WBRT (30 Gy in 10 fractions) with concurrent etoposide and cisplatin achieved a CR in less than half (44%) [18].
Although early and sufficient administration of steroids was required before administration of atezolizumab in the present case, Sorial et al. found no significant association between corticosteroid use before CIT and worse outcomes in patients with ES-SCLC or non-SCLC [14]. However, determining the efficacy of atezolizumab was difficult owing to the gap of 4.9 months between WBRT initiation and MRI re-evaluation; this is a significant limitation of this report.
The efficacy and safety of concurrent WBRT and CIT remain unclear. In the present case, CRs were sustained for 20 months for all BMs ≤17 mm; this indicates that microscopic BMs present at the initiation of WBRT can be completely eradicated by concurrent reduced-dose WBRT (<30 Gy) and CIT, which may attenuate WBRT-induced neurocognitive decline. In our case, chemo-WBRT followed by CIT resulted in remarkable regression of the large (>3.8 cm) BM but did not produce a CR. Residual viable tumors can eventually regrow and/or disseminate unless controlled by systemic therapy or additional EBRT. Therefore, focal dose escalation in the initial EBRT application likely improves local control, preferably achieving a CR, and a modest dose increase of ≥10%-20% may be sufficient. Integration of VMA with a simultaneous-integrated boost into WBRT enables focal dose escalation and dose reduction in affected and unaffected brain regions, respectively [1,8].
In the present case, no obvious progression of the thoracic lesions was evident until just before death; hence, death was attributed to CSF dissemination. Prevention of CSF dissemination is essential for improving the outcomes of patients with SCLC-BMs. Potential sources of CSF dissemination in the present case include the small tumor remnant fronting the regressed cavity of the left cerebellar lesion and the development of new BMs in the right caudate head and cerebellar surface, all of which contact the CSF space. Focal dose escalation in cases with large lesions would help eradicate one of the potential sources of dissemination. In cases with metachronous BMs, early detection and appropriate execution of salvage SRS is required before tumor seeding occurs.
To prevent or attenuate the development of new metachronous BMs, improving extracranial disease control is important. Consolidative thoracic radiotherapy for a partial response after CIT remains controversial [1,2,13,19]; however, some studies reported positive survival outcomes in patients who received this serial treatment [19]. Consolidative thoracic radiotherapy with a modest dose (e.g., 30 Gy) and image-guided VMA was a potential treatment option for the regressed thoracic lesions during the maintenance phase with atezolizumab in the present case.
Salvage re-irradiation of brainstem metastases developing after prior WBRT (30 Gy) is also challenging. The prescribed dose for salvage re-irradiation is commonly and frequently low owing to the risk of radiationinduced brainstem injury [15,20]. However, insufficient local control or tumor regrowth following SRS for brainstem metastases inevitably leads to significant impairment of neurological function, for which efficacious and safe salvage treatment options are extremely limited [15]. In the present case, sufficient GTV coverage with a BED 10 of 80 Gy in eight fractions and concurrent CPT-11 administration achieved CRs without ARE [15] for all treated lesions. However, due to the short observation period of 3.7 months, we could not conclude the validity of the treatment scheme.
Nevertheless, optimal strategies combining EBRT with systemic therapy for synchronous or metachronous SCLC-BMs in the CIT era should be reconsidered to further improve anti-tumor efficacy while preserving neurocognitive function [5,11,13].

Conclusions
Concurrent conventional WBRT without boost and EP administration followed by CIT can yield excellent and durable tumor responses in patients with SCLC-BMs, even those with BMs ≥3.8 cm. Using this regimen, a CR can be maintained for 20 months if the BM is ≤17 mm; however, it may be insufficient for BMs ≥3.8 cm.
In cases with large lesions, modest focal dose escalation should be considered to improve BM control and prevent CSF dissemination, along with dose de-escalation in the unaffected brain region to attenuate the decline of neurocognitive function. To prevent the development of new metachronous BMs, consolidative thoracic radiotherapy may be a potential treatment option for responders with active disease confined to the thorax following CIT.