The Impact of Surgery on Butterfly Gliomas – A Real-life Study of Biopsy versus Surgical Resection

Abstract

Introduction Although the extent of resection affects the overall survival of patients with gliomas, cytoreduction in patients with butterfly gliomas (BGs) remains controversial. Contemporary studies suggest that wide removal is feasible, but there is a lack of controlled, randomized, multicenter studies on the subject. The aim of the present study was to evaluate the impact of maximal safe resection (MSR) as a primary treatment strategy in patients with BG.

Methods A total of 30 consecutive patients, primarily treated with neurosurgery, were categorized into group A, in whom MSR was ≥ 50%, and group B, in whom MSR was < 50% (biopsy). The groups were studied and compared in terms of clinical and epidemiological characteristics, functionality, survival time (ST), and progression-free survival (PFS).

Results The results were analyzed through descriptive and inferential statistics. The most frequent tumor was IDH1-wild type glioblastoma, Grade 4 according to the World Health Organization (WHO) classification. Most patients (80%) had Karnofsky Performance Status (KPS) < 70%; however, there was no significant worsening in functionality associated with the interventions. Wider resections, i.e., those performed in group A, were associated with better Kaplan-Meier curves and ST compared with group B (68.35 versus 17.32 weeks respectively; p = 0.014). The PFS was also longer in group A compared to group B (57.70 versus 4.77 weeks, respectively; p = 0.012). In the multivariate analysis, resections > 50% were associated with reduced risk of recurrence (hazard ratio [HR] = 0.20; 95% confidence interval [95%CI] = 0.06–0.67; p = 0.009). The complication rates and functional decline were similar between groups.

Conclusion Surgical resections > 50% in patients with BG improved ST and PFS without a significant increase in morbidity or dysfunction.

Resumo

Introdução Embora a extensão da ressecção afetar a sobrevida geral dos pacientes com gliomas, a citorredução em pacientes com gliomas em asa de borboleta (GABs) permanece controversa. Apesar de estudos contemporâneos sugerirem que a remoção ampla é viável neste contexto, ainda faltam estudos controlados, randomizados e multicêntricos sobre o assunto. O objetivo deste estudo foi avaliar o impacto da máxima ressecção segura (MRS) como uma estratégia de tratamento primário em pacientes com GAB.

Métodos Ao todo, 30 pacientes consecutivos, tratados primariamente com neurocirurgia, foram categorizados como o grupo A, no qual a MRS foi ≥ 50%, e no grupo B, no qual a MRS foi < 50% (biópsia). Os grupos foram estudados e comparados quanto às características clínicas, epidemiológicas, funcionalidade, tempo de sobrevida (TS) e sobrevida livre de progressão (SLP).

Resultados Os resultados foram analisados por meio de técnicas de estatística descritiva e inferencial. O tumor mais frequente foi glioblastoma de tipo selvagem do gene IDH1, de grau 4 na classificação da Organização Mundial da Saúde (OMS). A maioria dos pacientes (80%) tinha Karnofsky Performance Status (KPS) < 70%, mas não houve piora significativa da funcionalidade associada às intervenções. Ressecções mais amplas, ou seja, as realizadas no grupo A, foram associadas a curvas de Kaplan-Meier e TS melhores em comparação com o grupo B (68,35 semanas versus 17,32 semanas, respectivamente; p = 0,014). A SLP também foi mais longa neste grupo em comparação com o grupo B (57,70 semanas versus 4,77 semanas, respectivamente; p = 0,012). Na análise multivariada, ressecções > 50% foram associadas a um risco reduzido de recorrência (razão de risco [RR] = 0,20; intervalo de confiança de 95% [IC95%] = 0,06–0,67; p = 0,009). As taxas de complicações e o declínio funcional foram semelhantes entre os grupos.

Conclusão Em pacientes com GAB, as ressecções cirúrgicas (> 50%) melhoraram o TS e a SLP, sem aumento significativo na morbidade e nas disfunções.

Keywords

butterfly glioma - biopsy - surgical resection - survival - progression-free survival

Palavras-chave

gliomas em asa de borboleta - biopsia - ressecção cirúrgica - sobrevida - sobrevida livre de progressão

Introduction

Diffuse gliomas rank among the most prevalent tumors of the central nervous system (CNS), with glioblastoma representing the third most common malignant brain tumor in adults (comprising 50.1% of the cases)[1]. A growth pattern that involves both cerebral hemispheres and the corpus callosum, referred to as butterfly glioma (BG), typically correlates with a grim prognosis, abbreviated survival, challenging management, cognitive decline, and swift functional deterioration.[2-5] Despite GB, World Health Organization (WHO) grade 4, is the histological subtype most frequently associated with this growth pattern, it can also manifest in astrocytomas WHO grade 2, 3, and 4, as well in lymphoproliferative disorders.

The role of cytoreduction as a primary treatment strategy cases of BG is still controversial.[5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] The involvement of eloquent areas and long fibers of critical association raises concerns about its indication, as it significantly increases patient morbidity and mortality.[5-7] Currently, improvements in surgical techniques and the growing number of evidence on the positive impact of surgical resection in the treatment of BG make physicians question the benefits of the more conservative management of the patients, including the strategy of biopsy followed by radiotherapy and chemotherapy.[5-18] The objective of the present study was to evaluate maximal safe resection (MSR) of the tumor used as a primary treatment strategy in a cohort of patients with BG, considering the survival time (ST) and the progression-free survival (PFS). The results were compared with those observed among patients treated in the same period with biopsy or removal of less than 50% of the total tumor volume (TTV). We also assessed the clinical and epidemiological characteristics of these patients, as well as the complications related to the hospital procedures.

Methods

Study Design and Inclusion Criteria

The present study was conducted through a careful retrospective analysis of information from medical records. We considered for inclusion all BG patients primarily treated with surgical resection who underwent adjuvant therapy (the Stupp protocol) at Instituto do Câncer do Estado de São Paulo Octávio Frias de Oliveira (ICESP) from January 2011 to August 2022.

Exclusion Criteria

We excluded all patients without radiological and histological confirmation of the diagnosis and those with severe diseases that could interfere with the ST, such as other malignant neoplasms, decompensated systemic diseases, and coronavirus disease 2019 (COVID-19).

Ethical Approval

The institutional Teaching and Research Commission and the Ethics Committee approved the study (under CCEP number: 1842/20). All living patients, or their legal representatives, signed the Free and Informed Consent Form to participate in the study.

Procedures

The participants were categorized into 2 groups based on the extent of the tumor removal: group A comprised patients with MSR ≥ 50% of the TTV, while group B included patients who underwent biopsy or MSR < 50% of the TTV. The choice of intervention was made during multidisciplinary discussions, considering the oncological and functional prognosis, as well as the impact and feasibility of the procedure. The clinical and epidemiological variables considered relevant included gender, race, age, symptoms, presence of hydrocephalus, involvement of three or more brain lobes (gliomatosis), overall health status, pre- and postoperative functional status, tumor histology, presence of the R132H mutation in the isocitrate dehydrogenase (NADP+) 1 (IDH1) gene (determined through immunohistochemistry), as well as PFS and ST. The general health conditions were evaluated using the classifications of the American Society of Anesthesiologists (ASA) and of the Eastern Cooperative Oncology Group (ECOG), and the functional status, through the Karnofsky Performance Status (KPS). For the statistical analysis, the time between surgery and the first radiological evidence of tumor volume increase was considered as the PFS. ST was defined as the time from surgery to death.

Statistical Analysis

We used the Jamovi (free, open source) and Stata (Statacorp. LLC, College Station, TX, United States) software for statistical analysis. The categorical variables were compared using the Chi-squared test, and they were expressed as absolute and relative frequencies. The normality of the continuous variables was assessed through Shapiro-Wilk tests. The parametric variables were expressed as means and standard deviation values and compared through the student's t-test. The non-parametric variables were evaluated using the Mann-Whitney U test. We used the long-rank test to compare survival curves developed through the Kaplan-Meier method. Univariate and multivariate Cox proportional hazards models were used to analyze possible variables associated with the ST. In the statistical analyses, values of p < 0.05 were considered significant.

Results

The current study included 30 patients equally distributed between groups A and B. The demographic, clinical, and epidemiological profiles of the sample are presented in [Table 1]. The mean age of the patients was 60.54(±11.43) years, with no statistically significant difference between the groups. In group B, there was a predominance of males (80.0%), and, in group A, the rate of male subjects was of 40% (p = 0.025).

Table 1 Clinical and demographical data of the study sample

		All	Group A*	Group B*	p-value
Cases: n (%)		30 (100.0)	15 (50.0)	15 (50.0)
Age in years: mean ± SD		60.54 ± 11.43	58.19 ± 8.73	61.65 ± 13.86	0.619
Sex: n (%)	Male	18 (60.0)	6 (40.0)	12 (80.0)	0.025
	Female	12 (40.0)	9 (60.0)	3 (20.0)
Comorbidities: n (%)	Arterial hypertension	10 (33.3)	5 (33.3)	5 (33.3)	1.000
	Diabetes mellitus	7 (23.3)	5 (33.3)	2 (13.3)	0.195
	Cardiovascular disease	3 (10.0)	2 (13.3)	1 (6.7)	0.543
	Stroke	2 (6.7)	2 (13.3)	0 (0.0)	0.143
	Smoking	5 (16.6)	3 (20.0)	2 (13.3)	0.510
Symptoms: n (%)	Headache	19 (63.3)	11 (73.3)	8 (53.3)	0.256
	Epilepsy	9 (30.0)	4 (26.7)	5 (33.3)	0.690
	Motor impairment	16 (53.3)	9 (60.0)	7 (46.7)	0.464
	Gait impairment	16 (53.3)	9 (60.0)	7 (46.7)	0.464
	Sensitive impairment	1 (3.3)	1 (6.7)	0 (0.0)	0.309
	Decreased sphincter control	13 (43.3)	7 (46.7)	6 (40.0)	0.713
	Decreased level of consciouness	12 (40.0)	8 (53.3)	4 (26.7)	0.136
	Delirium/Mental confusion	23 (76.7)	13 (86.7)	10 (66.7)	0.195
	Cognitive impairment	14 (46.7)	7 (46.7)	7 (46.7)	1.000
	Visual disturbance	5 (16.7)	3 (20.0)	2 (13.3)	0.624
ECOG: n (%)	≤ 2	12 (40)	6 (40)	6 (40)	1,000
	≥ 3	18 (60)	9 (60)	9 (60)
KPS: n (%)	≤ 60%	24 (80.0)	11 (73,3)	13 (86,7)	0.361
	≥ 70%	6 (20.0)	4 (26,7)	2 (13,3)
ASA score: : n (%)	≤ 2	10 (33.3)	8 (53.3)	2 (13.3)	0.020
	≥ 3	20 (66.7)	7 (46,7)	13 (86.7)
Histology: n (%)	Grade-2 astrocytoma	1 (3.3)	0 (0.0)	1 (6.7)	0.510
	Grade-3 astrocytoma	3 (10.0)	2 (13.3)	1 (6.7)
	Grade-4 glioblastoma	26 (86.7)	13 (86.7)	13 (86.6)
IDH1 mutation: n (%)	Mutation	1 (3.3)	0 (0.0)	1 (6.7)	0.030
	Non-mutation	16 (53.4)	5 (33.3)	11 (73.3)
	NOS	13 (43.3)	10 (66.7)	3 (20.0)
Hydrocephalus: n (%)		13 (43.0)	6 (46.2)	7 (53.8)	0.713
Gliomatosis: n (%)		22 (73.3)	10 (45.5)	12 (54.5)	0.409
Survival in weeks: mean ± SD		42.84 ± 65.16	68.35 ± 29.57	17.32 ± 29.57	0.014
Recurrence in weeks: mean ± SD		30.23 ± 70.41	57.70 ± 93.97	4.77 ± 7.15	0.012

Abbreviations: ASA, American Society of Anesthesiologists physical status classification; ECOG, Eastern Cooperative Oncology Group; KPS, Karnofsky Performance Scale; SD, standard deviation.

The most frequent symptoms reported uniformly in both groups, were delirium/mental confusion (76.7%), headache (63.3%), motor alterations (53.3%), and gait alterations (53.3%). Loss of sphincter control and epilepsy were reported by 43.3% and 30% of the patients, respectively. There were no statistically significant differences regarding the presence of hydrocephalus or the pattern of gliomatosis growth at presentation. The groups were similar regarding the most prevalent comorbidities, which included hypertension (33.3%), diabetes mellitus (23.3%), and cardiovascular disease (10%). They were also similar when categorized by functional performance, considering the KPS (≥ 70% and < 70%) and ECOG (≥ 3 and < 3) scales. Despite this, patients in group B were in better general health conditions (ASA < 3) at the time of the intervention (p = 0.02).

Malignant gliomas, WHO grades 3 and 4, were predominant and represented 86.7% and 10% of the cases respectively. Only 1 patient (3.3%) was classified as presenting diffuse astrocytoma, WHO grade 2, mutated IDH1. The surgical mortality rate (death up to the seventh day of the intervention) was of 0%. In total, 2 patients (13.33 %) from group A and one patient (6.66 %) from group B showed functional deterioration related to the intervention (p > 0.05).

Survival Time

The mean ST was higher in group A compared to group B (68.35 vs. 17.32 weeks, respectively; p = 0.014), and the statistically significant differences were maintained in the univariate analysis (hazard ratio [HR] = 0.39; 95% confidence interval [95% CI] = 0.18–0.85; p =  0.018) and multivariate analysis (HR = 0.22; 95% CI = 0.07–0.69; p = 0.010) models used ([Table 2]). A comparison of the Kaplan-Meier survival curves is shown in [Figure 1].

Table 2 Cox regression for the survival time

	Univariate analysis			Multivariate analysis
	HR	95%CI	p-value	HR	95%CI	p-value
Sex (male)	0.65	0.30–1.39	0.262	0.65	0.24–1.73	0.382
Age (< 60 years)	1.73	0.79–3.76	0.168	2.33	0.82–6.62	0.113
ASA score (< 3)	0.95	0.43–2.08	0.890	5.06	1.33–19.29	0.017
KPS (≥ 70%)	0.38	0.14–1.02	0.054	0.63	0.19–2.13	0.460
Surgery	0.39	0.18–0.85	0.018	0.22	0.07–0.69	0.010

Abbreviations: 95%CI, 95% confidence interval; ASA, American Society of Anesthesiologists physical status classification; HR, hazard ratio; KPS, Karnofsky Performance Scale; SD, standard deviation.

Fig. 1 Kaplan-Meyer analysis of the survival time.

The patient who achieved the longest survival underwent surgical excision of a glioblastoma, IDH1-wild type, WHO grade 4. In addition to gross total resection, the patient received comprehensive adjuvant therapy following Stupp's protocol. Remarkably, this individual remains alive with a Karnofsky Performance Status of 90 % and has been free of recurrence for 6 years and 7 months of follow-up ([Figure 2]).

Fig. 2 Illustrative case. Butterfly Glioblastoma, IDH1-wild type, WHO grade 4.

Progression Free-Survival

The mean time until recurrence was 30.23 (±70.41) weeks. The PFS was significantly higher in group A compared to group B (57.70 versus 4.77 weeks respectively; p = 0.012) and differences in the Kaplan-Meier curves for recurrence are shown in [Figure 3]. After 25 weeks of follow-up, all patients in group B, who were treated with biopsy, had already presented recurrence, compared to 60 % of group A. A statistically significant reduction in the risk of recurrence was observed only in group A, both in the univariate (HR = 0.35; 95% CI = 0.15–0.79; p = 0.011) and multivariate (HR = 0.20; 95% CI = 0.06–0.67; p = 0.009) analyses ([Table 3]).

Fig. 3 Kaplan-Meyer analysis of the progression-free survival.

Table 3 Cox regression for progression-free survival

	Univariate analysis			Multivariate analysis
	HR	95%CI	p-value	HR	95%CI	p-value
Sex (male)	0.54	0.25–1.18	0.121	0.51	0.19–1.40	0.192
Age (< 60 years)	1.94	0.88–4.29	0.100	2.02	0.71–5.75	0.190
ASA score (< 3)	0.63	0.28–1.43	0.274	4.11	1.00–16.93	0.050
KPS (≥ 70%)	0.51	0.19–1.36	0.180	0.74	0.22–2.56	0.638
Surgery	0.35	0.15–0.79	0.011	0.20	0.06–0.67	0.009

Abbreviations: 95%CI, 95% confidence interval; ASA, American Society of Anesthesiologists physical status classification; HR, hazard ratio; KPS, Karnofsky Performance Scale; SD, standard deviation.

Discussion

Butterfly gliomas are commonly characterized by aggressive behavior, unfavorable histological and molecular profiles, short STs, early recurrence, and poor prognosis, regardless of treatment. Nevertheless, many cases require surgical intervention due to rapid progression, for symptom relief, or simply to establish a histological diagnosis. Although historically MSR has been infrequently indicated to treat BG, there are compelling reasons to consider it. Adequate surgical removal, when combined with adjuvant therapies such as radiotherapy and chemotherapy, can substantially extend survival and enhance the quality of life of patients. This is achieved by reducing intracranial pressure, alleviating pressure on the surrounding brain structures, and potentially mitigating neurological symptoms such as headaches, motor and cognitive deficits, and seizures.

Consistent with the existing literature, the current study predominantly featured gliomas classified as WHO grades 3 and 4 of astrocytic origin, with wild type IDH1 status. Only 1 case was categorized as diffuse astrocytoma, mutant IDH-1, WHO grade 2. Despite this, we noted a significant improvement in both ST and PFS among patients in whom surgical resection exceeded 50% of the TTV group A. Our analysis of the clinical and epidemiological profile highlighted the advanced mean age, functional impairment, and overall frailty of the affected population. Surprisingly, 66.7% and 80% of the cases were classified as ASA ≥ 3 and KPS < 70 respectively. Noteworthy comorbidities included hypertension, diabetes mellitus, cardiovascular diseases, and a history of smoking. The most common symptoms reported were headache, epileptic seizures, and frontal and psychiatric manifestations (delirium, motor changes, cognitive impairments, and loss of sphincter control). The demographic profile, combined with the disease's aggressive nature, justifies the apprehension many neurosurgeons experience when managing these cases and opting for more conservative interventions. However, in the present study, the increase in ST and PFS observed in group A was not associated with higher rates of surgical morbidity, mortality, or functional impairment resulting from the intervention. The effect was particularly pronounced in PFS, which was extended by more than 12 times. Notably, surgery emerged as an independent factor, even in statistical models incorporating established factors such as age, IDH1 status, and functional performance. These findings support recent literature suggesting the favorable impact of MSR as a primary treatment modality in such cases.[3] [4] [5] [9] [10] [11] [13] [15] [16] [17] [18]

Despite technological advances, established and described techniques for the resection of these tumors, and along with evidence favorable to this strategy,[4] [9] [11] surgery is still seldom used in the primary treatment of BG. This may be due to poor results reported in the past and the lack of know-how and expertise among neurosurgeons in this type of intervention. Despite the infiltrative and diffuse character of these tumors, we believe that most of the symptoms, dysfunctions, and mechanisms of death are due to edema, distortion of CNS areas, and fibers.

A recent study[18] found that GBs can increase in volume by an average of 1.1% per day, potentially doubling in size in just 22 days. Analogous to the management of other tumors, as well as gliomas affecting different parts of the CNS, we believe that preserving the quality of life in BG patients and reversing or delaying dysfunctions and causes of death may be achieved through more extensive surgical resections. However, definitive conclusions on this matter would require better designed and randomized studies. Yet, the rarity of the disease and ethical considerations pose challenges, potentially rendering such studies unfeasible. Real-world data, such as those presented herein, are vital to support evidence-based practices without depriving functionally sound patients of optimal treatment options or wasting resources on those with poor prognoses. Nevertheless, the current study has notable limitations, including its retrospective design, the single-center design, sample size, molecular data availability, patient inclusion period, and estimation of TTV.

Conclusion

Surgically removing at least 50% of a BG has emerged as a positive independent factor linked to a more favorable outcome in terms of ST and PFS. This improvement was not accompanied by elevated surgical morbidity and mortality rates. However, the present study did not identify specific subgroups that might benefit the most from this approach. These findings align with those of recent research[3] [4] [5] [9] [10] [11] [13] [15] [16] [17] [18] suggesting the superiority of extensive surgical resection over primary management with biopsy alone. Nonetheless, further well-designed, multicenter, randomized studies are still required to reach definitive conclusions.

Conflict of Interests

The authors have no conflict of interests to declare.

• References

• 1 Ostrom QT, Price M, Neff C. et al. CBTRUS Statistical Report: Primary Brain and Other Central Nervous System Tumors Diagnosed in the United States in 2015-2019. Neuro-oncol 2022; 24 (5, Suppl 5): v1-v95
• 2 Palmisciano P, Ferini G, Watanabe G. et al. Gliomas Infiltrating the Corpus Callosum: A Systematic Review of the Literature. Cancers (Basel) 2022; 14 (10) 2507
• 3 Dziurzynski K, Blas-Boria D, Suki D. et al. Butterfly glioblastomas: a retrospective review and qualitative assessment of outcomes. J Neurooncol 2012; 109 (03) 555-563
• 4 Bjorland LS, Dæhli Kurz K, Fluge Ø. et al. Butterfly glioblastoma: Clinical characteristics, treatment strategies and outcomes in a population-based cohort. Neurooncol Adv 2022; 4 (01) vdac102
• 5 Tunthanathip T, Ratanalert S, Sae-Heng S, Oearsakul T. Butterfly tumor of the corpus callosum: Clinical characteristics, diagnosis, and survival analysis. J Neurosci Rural Pract 2017; 8 (5, Suppl 1): S57-S65
• 6 Molinaro AM, Hervey-Jumper S, Morshed RA. et al. Association of Maximal Extent of Resection of Contrast-Enhanced and Non-Contrast-Enhanced Tumor With Survival Within Molecular Subgroups of Patients With Newly Diagnosed Glioblastoma. JAMA Oncol 2020; 6 (04) 495-503
• 7 Li YM, Suki D, Hess K, Sawaya R. The influence of maximum safe resection of glioblastoma on survival in 1229 patients: Can we do better than gross-total resection?. J Neurosurg 2016; 124 (04) 977-988
• 8 Wang L, Liang B, Li YI, Liu X, Huang J, Li YM. What is the advance of extent of resection in glioblastoma surgical treatment-a systematic review. Chin Neurosurg J 2019; 5 (01) 2
• 9 Franco P, Delev D, Cipriani D. et al. Surgery for IDH1/2 wild-type glioma invading the corpus callosum. Acta Neurochir (Wien) 2021; 163 (04) 937-945
• 10 Chaichana KL, Jusue-Torres I, Lemos AM. et al. The butterfly effect on glioblastoma: is volumetric extent of resection more effective than biopsy for these tumors?. J Neurooncol 2014; 120 (03) 625-634
• 11 Chojak R, Koźba-Gosztyła M, Słychan K. et al. Impact of surgical resection of butterfly glioblastoma on survival: a meta-analysis based on comparative studies. Sci Rep 2021; 11 (01) 13934
• 12 Siddiqui J, Krishnan AS. Butterfly Glioma. N Engl J Med 2018; 378 (03) 281-281
• 13 Beaumont TL, Mohammadi AM, Kim AH, Barnett GH, Leuthardt EC. Magnetic resonance imaging-guided laser interstitial thermal therapy for glioblastoma of the corpus callosum. Neurosurgery 2018; 83 (03) 556-565
• 14 Burks JD, Bonney PA, Conner AK. et al. A method for safely resecting anterior butterfly gliomas: the surgical anatomy of the default mode network and the relevance of its preservation. J Neurosurg 2017; 126 (06) 1795-1811
• 15 Opoku-Darko M, Amuah JE, Kelly JJP. Surgical Resection of Anterior and Posterior Butterfly Glioblastoma. World Neurosurg 2018; 110: e612-e620
• 16 Dayani F, Young JS, Bonte A. et al. Safety and outcomes of resection of butterfly glioblastoma. Neurosurg Focus 2018; 44 (06) E4
• 17 Chawla S, Kavouridis VK, Boaro A. et al. Surgery vs. Biopsy in the Treatment of Butterfly Glioblastoma: A Systematic Review and Meta-Analysis. Cancers (Basel) 2022; 14 (02) 314
• 18 Müller DMJ, De Swart ME, Ardon H. et al. Timing of glioblastoma surgery and patient outcomes: a multicenter cohort study. Neurooncol Adv 2021; 3 (01) vdab053

Publication History

Received: 18 June 2024

Accepted: 21 October 2024

Article published online:
27 March 2025