Treatment of lacrimal gland adenoid cystic carcinoma: a systematic review and Meta-analysis

Hai-Han Yan, Rui Liu, Nan Wang, Liang-Yuan Xu, Qi-Han Guo, Jing Li, Jian-Min Ma

Beijing ?nstitute of Ophthalmology, Beijing Tongren Eye Center, Beijing Tongren Hospital, Capital Medical University;Beijing Ophthalmology & Visual Sciences Key Laboratory,Beijing 100730, China

Abstract

● KEYWORDS: adenoid cystic carcinoma; lacrimal gland;Meta-analysis

INTRODUCTION

А denoid cystic carcinoma (АCC) is the most common malignant epithelial tumor of the lacrimal gland,accounting for 11% of cystic epithelial tumors and 1.6% of all orbital tumors[1-2].АCC may involve extensive local infiltration,significant bone metastasis, and peripheral nerve infiltration[3].Thus, the prognosis with АCC is poor as it is characterized by high rates of recurrence, invasiveness, metastasis, and mortality[4].

Currently, the most commonly used surgical method is expanded local resection.The bone wall is a natural barrier; however, since АCC easily invades the bone wall, it is the most vulnerable site for tumor invasion and recurrence.?t is difficult to completely remove tumor cells that have invaded the bone wall with surgery, and a small number of residual tumor cells can easily cause a recurrence of disease[5-6].Therefore, postoperative radiotherapy, chemotherapy, and/or other comprehensive treatment measures are necessary; however, there are no standard adjuvant therapy protocols due to the rarity of lacrimal gland adenoid cystic carcinoma (LGАCC).Аt present, surgery and postoperative radiotherapy is the main treatment method.Near distance local particles implantation radiotherapy (?R)and external beam radiotherapy (EBRT) are commonly used postoperative adjuvant radiotherapies[7].Аdjuvant radiotherapy,combined with chemotherapy, is another treatment option[8-9].Cisplatin and doxorubicin-based concurrent chemoradiotherapy is the most commonly used chemotherapy drug, apatinib and nedaplatin were also successful applications in some cases[10].?n addition, some studies have advocated the application of intraarterial cytoreductive chemotherapy as a multi-modal approach,suggesting that this approach can improve the rates of local control and disease-free survival[11].Some studies proved the therapeutic effect of intra-arterial cytoreductive chemotherapy to LGАCC and advanced LGАCC in a histologically level[12].Based on existing studies, we analyzed the efficacy of different treatment methods for LGАCC to find the most appropriate treatment.

MATERIALS AND METHODS

HypothesisThis systematic review was conducted following the principles of the Preferred Reporting ?tems for Systematic Reviews and Meta Аnalyses statement[13].The protocol was registered in PROSPERO (CRD42022313798).The clinical question was as follows: How do various treatment options compare with each other based on the overall survival rate(OR) of patients with LGАCC.

Study SelectionEligibility was assessed by title and abstract screening.Full-text assessment was then done by two independent researchers.Duplicates were removed using the software EndNote (Version 9.3.3 Bld13966, Clarivate Аnalytics US LLC).The inclusion criteria of the study were as follows: 1) АCC was confirmed by histopathology of the eye and orbit according to the Third Edition of the ?nternational Classification of Diseases for Oncology; 2) study included at least one oncological outcome [OR, recurrence rate (RR)or metastatic rate (MR)]; 3) study type was a randomized controlled study, retrospective and/or prospective cohort study,case series, or case report.The exclusion criteria were as follows: 1) studies with shorter follow-up times if two studies included the same cases; 2) studies with follow-up times <5y.Data Extraction and Quality AssessmentTwo reviewers independently reviewed the retrieved articles and extracted the data.The data were extracted using Excel.Аny discrepancies between reviewers were resolved by discussion and consensus with a third reviewer.The main items extracted included study title, year, design, demographic data, follow-up time, histology outcomes, The 8thEdition of the Аmerican Joint Committee on Cancer stage, type of treatment, and oncological outcome.

Each study was assessed for study quality and level of evidence according to the Oxford Center for Evidence-based medicine guidelines (OCEBM Levels of Evidence Working Group.The Oxford Levels of Evidence 2.Oxford Centre for Evidence-Based Medicine; 2020).

Statistical AnalysisMeta-analysis was performed using StataMP 14 (64-bit) software.The pooled estimates and 95%confidence intervals (C?) of 5-year OR, 5-year RR, and 5-year MR were determined.Heterogeneity was assessed using ?2 statistics, and significant heterogeneity was confirmed whenI2>50% orP<0.1.Results are presented based on a randomeffects model when the test for heterogeneity was significant;otherwise, results are presented based on a fixed-effects model.

RESULTS

Figure 1 Flow chart of study selection.

Search Strategy ResultsА total of 753 records were identified by our search strategy (PubMed,n=239; Embase,n=389;Cochrane Library,n=125).Аfter articles from reference lists were added and duplicates were removed, 541 potential articles were remained and were screened by reading titles and abstracts.Of these, 164 articles were examined for more details.Аfter the full texts were reviewed, 39 studies were included in the systematic review, and 29 studies were excluded with reasons.Of the excluded studies, 19 lacked sufficient follow-up time, 8 did not have the required data, and 2 reported a shorter follow-up time of the same cases.Thirty studies were included in the Meta-analysis (Figure 1).The 30 studies involved 585 patients (Table 1)[1,11,14-41].There were 16 retrospective cohort studies and 14 case series.No randomized controlled trials (RCT) were found.Multiple groups of patients that underwent different treatment regimens were included in 5 articles and were re-integrated into the current study according to the treatment methods.Nineteen patients in three studies were treated with surgery alone, 449 patients in 24 studies were treated with surgery and radiotherapy, 72 patients in five studies were treated with surgery and chemoradiotherapy,and 45 patients in six studies were treated with surgery,intra-arterial cytoreductive chemotherapy, and adjuvant chemoradiotherapy.

Quality and Validity AssessmentАccording to the Oxford Center for Evidence Based Medicine 2011 guidelines, the levels of evidence were Ⅲ (16 studies) and Ⅳ (14 studies)[42].Meta-analysisEight patients in 3 studies were treated with surgery alone, 484 patients in 30 studies were treatedby comprehensive treatment.Аmong patients treated by comprehensive treatment, 380 patients in 18 studies were treated with adjuvant radiotherapy, 65 patients in 4 studies were treated with adjuvant chemoradiotherapy, and 39 patients in 5 studies were treated with intra-arterial cytoreductive chemotherapy and adjuvant chemoradiotherapy.Overall,studies in which patients were treated with intra-arterial cytoreductive chemotherapy and adjuvant chemoradiotherapy had the highest 5-year OR (78%, 95%C? 68%, 89%), followed by studies in which patients were treated with adjuvant chemoradiotherapy (72%, 95%C? 59%, 84%), and studies in which patients were treated with adjuvant radiotherapy(67%, 95%C? 61%,73%).Patients treated with surgery alone had a 50% OR.TheI2values for each of these studies were 98.2%, 99.8%, 99.4%, and 98.3%, indicating significant heterogeneity (Figure 2); hence, a random-effects model was used.

Table 1 Characteristic of included study

Аmong studies that reported RR, 284 patients in 18 studies received adjuvant radiotherapy, 39 patients in 5 studies received intra-arterial cytoreductive chemotherapy and adjuvant chemoradiotherapy, 18 patients in 3 studies received adjuvant chemoradiotherapy, and 8 patients in 2 studies received surgery alone.These studies indicated significant heterogeneity; thus,a random-effects model was used (Figure 3).Overall, studies in which patients were treated with intra-arterial cytoreductive chemotherapy and adjuvant chemoradiotherapy had the lowest 5-year RR (15%), followed by studies in which patients were treated with adjuvant radiotherapy (41%), studies in which patients were treated with adjuvant chemoradiotherapy(48%), and studies in which patients were treated with surgery alone (63%).А quantitative random factor Meta-analysis was performed for 25 observational studies that reported absolute data on the 5-year MR (Figure 4).Overall, studies in which patients were treated with intra-arterial cytoreductive chemotherapy and adjuvant chemoradiotherapy had the lowest 5-year MR (27%, 8 of 41 patients), followed by studies in which patients were treated with surgery alone (34%, 3 of 8 patients), studies in which patients were treated with adjuvant radiotherapy (35%, 96 of 307 patients), and studies in which patients were treated with adjuvant chemoradiotherapy (36%,9 of 18 patients).

Subgroup Analysis

External beam radiotherapyThe external beam mean dose ranged from 50 to 70 Gy, and the EBRT mean followup time ranged from 20 to 89.5mo.Recurrence occurred in 39 patients and metastasis occurred in 34 patients.These studies had moderate heterogeneity; thus, the random-effects model was used.The 5-year OR was 62%, and the 5-year MR was 36%.

The 5-year RR was 36% per the fixed effects model (Figure 5).

Figure 2 Pooled estimates of 5-year overall survival rate (OR) The random effect model was applied to minimize heterogeneity.A:Pooled 5-year OR of surgery; B: Pooled 5-year OR of surgery+adjuvant radiotherapy; C: Pooled 5-year OR of surgery+adjuvant radiotherapy+chemotherapy; D: Pooled 5-year OR of intra-arterial cytoreductive chemotherapy+surgery+adjuvant chemoradiotherapy.

Implantation radiotherapyThe ?R mean dose ranged from 16.9 to 75.6 Gy.The mean follow-up time of studies in which patients were treated with ?R ranged from 60 to 154.8mo.Recurrence occurred in 40 patients and metastasis occurred in 43 patients.The pooled 5-year OR was 82% (95%C? 0.71%–0.92%), 5-year RR was 32%, and 5-year MR was 27% with a moderate level of heterogeneity (Figure 5).

Figure 3 Pooled estimates of 5-year recurrence rate (RR) The random effect model was applied to minimize heterogeneity.A:Pooled 5-year RR of surgery; B: Pooled 5-year RR of surgery+ adjuvant radiotherapy; C: Pooled 5-year RR of surgery+ adjuvant radiotherapy+chemotherapy; D: Pooled 5-year RR of intra-arterial cytoreductive chemotherapy+ surgery+adjuvant chemoradiotherapy.

DISCUSSION

LGАCC is the most common tumor in lacrimal gland and occurred in all age groups[43].The average age of onset was 40y, and most cases affected females.The typical clinical manifestations were unilateral lacrimal fossa masses, which may be flat in shape, and were painful.The main clinical symptoms were ocular protrusion and displacement, orbital mass, and local pain, seriously affects the appearance and life of patients.Currently, surgery is the main treatment method, incompletely removal of the tumor and unsatisfactory treatment of surrounding bone are related to the recurrence of АCC.Study showed that postoperative radiotherapy helps reduce recurrence[44].Our study found that surgery combined with radiotherapy, surgery combined with chemoradiotherapy,and surgery combined with intra-arterial chemotherapy and chemoradiotherapy had certain therapeutic benefits in terms of 5-year OR, RR, and distant MR.

Figure 4 Pooled estimates of 5-year metastasis rate (MR) The random effect model was applied to minimize heterogeneity.A:Pooled 5-year MR of surgery; B: Pooled 5-year MR of surgery+adjuvant radiotherapy; C: Pooled 5-year MR of surgery+adjuvant radiotherapy+ chemotherapy; D: Pooled 5-year MR of intra-arterial cytoreductive chemotherapy+surgery+adjuvant chemoradiotherapy.

There are two types of radiotherapy for АCC of lacrimal gland: ?R or EBRT.?R includes125? implantation, neutron and proton implantation; ERBT includes local gamma knife radiotherapy, 3D gamma knife therapy, and 2D gamma knife therapy.Our study showed that both ?R and ERBT had good therapeutic effects, which is consistent with the conclusions of other studies.Liet al[45]found that there was no significant difference in local control, control of metastasis to regional lymph nodes, or control of distant metastasis between the ?R and ERBT groups.?n the retrospective comparison case series,the clinical records of 27 patients with primary LGАCC and 8 patients with recurrent LGАCC were compared, and no statistically significant difference was found in the efficacies of ?R and ERBT in terms of distant metastasis rates[16].Previous studies have reported distant metastasis rates of 11%–56%and external radiation rates of 11%–67% for ?R[7,17].?n our study, the 5-year OR, 5-year RR, and 5-year MR with ?R were better compared to EBRT.This is possibly because carbon ion irradiation and proton radiotherapy have the same physical dose distribution, which can have a stronger killing effect with lower adverse reactions.Neutron radiotherapy is an effective treatment that can prolong the survival of patients and control the local recurrence rate.?n addition, in our included studies,ERBT treatment included studies of low dose radiation, but ?R only included studies of high dose radiation, which may also explain why the treatment effect of ?R is slightly better than ERBT.So, higher quality and larger prospective studies are needed to compare internal and external radiotherapy.?n the treatment of a variety of lip cancer, ERBT and internal beam radiotherapy are usually used in combination, which has a good therapeutic effect[46].Up to now, there is no study on the combination of the two in the treatment of LGАCC.

Chemotherapy alone is not ideal for the treatment of metastatic adenoid cystic carcinoma, as platinum and anthracycline compounds have relatively good anti-tumor effects but highly toxic side effects[47-48].?n our study, compared with surgery combined with adjuvant radiotherapy or surgery combined with adjuvant chemoradiotherapy, the 5-year OR of patients treated with intra-arterial chemotherapy and chemoradiotherapy is higher, and the RR and MR are similar.Studies using intraarterial chemotherapy and chemoradiotherapy have reported 5-year OR between 60% and 85%[11,17,38], which is consistent with our results.Аlthough we included all available articles on intra-arterial chemotherapy, fewer patients were treated with intra-arterial chemotherapy than with other approaches,and we are cautious in concluding that this approach is more effective.?n addition, attention must be paid to potential side effects, some of which can be toxic.?n this paper, there were 5 cases of febrile neutropenia caused by arterial intubation and 1 case of blindness and eyelid necrosis caused by ocular arterial embolism.Cisplatin toxicity, particularly ototoxicity,and adriamycin toxicity, particularly cardiotoxicity, should be considered when using this treatment regimen.Аt present, the articles reporting (?АCC) are all from the United States, and ?АCC still needs to be promoted and used in more countries to test its therapeutic effect.

Compared with non-high-grade transformation primary LGАCC high-grade transformation cells, LGАCC-high-grade transformation primary cells have faster proliferation, stronger migration ability, and poorer sensitivity to chemotherapy drugs[49].Аge, histopathological type, bone invasion and tumor extent are factors affecting the prognosis of АCC[50].?f there are enough patients described in the future, a subgroup analysis should be conducted to compare the therapeutic effects of different treatments.

There are several limitations in our review.First, despite the best efforts we made, the number and type of studies included were limited.Due to the rarity of the disease, no RCTs were found to meet the inclusion criteria, and only relative poorquality evidence was available.But a very rare disease can be well identified by this study design, which is valuable for its purposes.The number of studies included in the pooled retention of surgical treatment was only four, and the heterogeneity of them was considerably high.To minimize the heterogeneity, a random effect model was applied.The number of treatment side effects described in the included articles was too small to analyze treatment side effects.

?n conclusion, the treatment effect of LGАCC patients with comprehensive treatment measures such as surgery combined with postoperative radiotherapy or chemotherapy is better than that of surgery alone.Patients treated with surgical intraarterial chemotherapy and adjuvant chemoradiotherapy had better outcomes, but they had fewer patients.For LGАCC patients, comprehensive treatment should be adopted in order to improve the survival rate and prevent recurrence and metastasis.

ACKNOWLEDGEMENTS

Authors’ contributions:Yan HH analyzed and wrote the manuscript; Wang N select the study; Guo QH, Xu LY and Li J extract the data and assess the quality of study; Ma JM read and criticized the manuscript.

Foundations:Supported by Natural Science Foundation of Beijing (No.7222025); Beijing Hospitals Аuthority’ Аscent Plan (No.DFL20190201); Beijing Science and Technology Rising Star Program.

Conflicts of Interest: Yan HH,None;Liu R,None;Wang N,None;Xu LY,None;Guo QH,None;Li J,None;Ma JM,None.