Key Points
Question
What is the risk of extraocular extension from injecting chemotherapy into eyes with retinoblastoma?
Findings
This 10-center cohort study, with an accumulated 3553 injections in 655 patients, identified no events of extraocular tumor that could be attributed to prior intravitreous chemotherapy injections.
Meaning
These data suggest that the occurrence of extraocular tumor in eyes with retinoblastoma receiving intravitreous chemotherapy is possible but unlikely.
Abstract
Importance
The risk of extraocular extension from injecting chemotherapy into eyes with retinoblastoma is minimally understood; however, understanding this risk is important because of the increasing use of intravitreous chemotherapy.
Objective
To evaluate the risk of extraocular extension in eyes with retinoblastoma that have received intravitreous chemotherapy injections.
Design, Setting, and Participants
This retrospective cohort study was performed in 655 patients at 10 retinoblastoma centers in North and South American, European, Israeli, and Chinese centers. Physicians at the retinoblastoma centers administered more than 120 intravitreous chemotherapy injections in eyes with retinoblastoma from February 1, 1999, through February 28, 2017.
Main Outcomes and Measures
Risk of extraocular extension with secondary observational variables, including injection and precautionary techniques.
Results
A total of 3553 intravitreous chemotherapy injections (3201 melphalan hydrochloride, 335 topotecan hydrochloride, and 17 methotrexate sodium) were administered to 704 eyes in 655 patients with retinoblastoma (mean [SD] age of patients at the time of the initial injections, 31.6 [11.6] months; 348 male [53.1%]). There were no extraocular tumor events related to prior intravitreous injections. This finding resulted in a calculated proportion of zero extraocular events per eye. According to the rule of 3, the risk is no greater than 0.08% injections. All 10 centers included in this study used at least 2 presumed precautionary injection methods (lowering of intraocular pressure, cryotherapy, ocular surface irrigation, ultrasonic biomicroscopy surveillance of the injection site, and subconjunctival chemotherapy deposition).
Conclusions and Relevance
With use of at least 2 presumed precautionary safety methods, no extraocular extension of tumor events occurred. According to the rule of 3, this finding suggests that the risk is no greater than 0.08% injections.
This study evaluates the risk of extraocular extension in eyes with retinoblastoma that have received intravitreous chemotherapy injections.
Introduction
Historically, there has been concern that breaching the wall of an eye with retinoblastoma can result in complications for the eye and the patient. Aspiration techniques in eyes with retinoblastoma have demonstrated tumor within the needle tract.1 Furthermore, biopsy or intraocular surgery on these diseased eyes has reportedly resulted in reactivation of tumor necessitating enucleation or leading to orbital and/or systemic metastases.2
In the past, these fears hindered the acceptance of placing any needle, including for drug injection into the vitreous, into an eye with retinoblastoma. However, with the adoption of safety-enhanced techniques, the risk of tumor externalization with intravitreous chemotherapy injections is thought to be low. A meta-analysis by Smith and Smith3 in 2013 calculated the proportion of patients with extraocular spread of tumor to be 0.007. That study was restricted to a literature review of studies performed during 53 years between 1960 and 2013, with details limited to those reported in published work. The current study analyzed the current risk of extraocular dissemination of tumor by evaluating more than 3500 intravitreous chemotherapy injections in 655 patients that were performed at high-volume retinoblastoma centers worldwide, with most performed during the past 10 years.
Methods
This study was performed at 10 North and South American, European, Israeli, and Chinese centers with physicians who were members of the International Society of Ocular Oncology who had administered more than 120 intravitreous chemotherapy injections in eyes with retinoblastoma from February 1, 1999, through February 28, 2017. Medical records were retrospectively reviewed to determine details on the number of patients, eyes, and injections. Details were collected regarding the needle size; injection site; drug injected; injection technique; and presumed safety-enhancement measures, including lowering of intraocular pressure (IOP), cryotherapy, ocular surface irrigation, ultrasonographic biomicroscopy surveillance of the injection site, and subconjunctival chemotherapy deposition. The study specifically evaluated the number of injections that were associated with extraocular dissemination of tumor at the needle site. This study was approved by the institutional review board of Memorial Sloan Kettering Cancer Center. Written informed consent was obtained before the intravitreous injection procedure, in the presence of a translator if required. All data were deidentified.
Results
A total of 3553 intravitreous chemotherapy injections (3201 melphalan hydrochloride, 335 topotecan hydrochloride, and 17 methotrexate sodium) were administered to 704 eyes in 655 patients with retinoblastoma (mean [SD] age of patients at the time of the initial injections, 31.6 [11.6] months; 348 male [53.1%]). The mean number of intravitreous chemotherapy injections per eye was 4.5. Each eye had a follow-up time after the last intravitreous injection of at least 6 months (the longest follow-up was 98 months). No extraocular tumor events were related to intravitreous injection. According to the rule of 3 (ie, if none of n patients has the adverse event in question, there is reasonable confidence [95%] that the true rate of this event in the population is no more than 3 in n [3/n]),4,5 the 95% confidence of a calculated risk of extraocular extension was no greater than 0.08% injections.
Numerical details regarding intravitreous injections at each center are given in Table 1. Across the centers, needle size ranged from 30 to 33 in gauge and 9 to 12 mm in length. Injection site from the limbus ranged from 2.0 to 3.5 mm, and drug volume ranged from 0.04 to 0.20 mL. The dose of injected melphalan hydrochloride ranged from 20 to 40 μg, and if used, 20 μg of topotecan hydrochloride and 400 of μg methotrexate were given.
Table 1. Numerical Details Regarding Intravitreous Injections at Each Center.
| Variable | United States (New York, NY) | Switzerland (Lausanne) | China (Beijing) | United States (Philadelphia, PA) | Brazil (São Paulo) | United States (Houston, TX) | France (Paris) | Italy (Sienna) | United States (Los Angeles, CA) | Israel (Jerusalem) |
|---|---|---|---|---|---|---|---|---|---|---|
| Time of first intravitreous injection | Sep 2012 | Jan 2009 | Aug 2013 | Oct 2007 | Nov 2012 | Nov 2014 | Feb 1999 | Jul 2012 | Dec 2012 | Dec 2011 |
| No. of patients | 141 | 124 | 98 | 78 | 54 | 35 | 38 | 31 | 30 | 26 |
| No. of eyes | 152 | 131 | 107 | 82 | 55 | 43 | 39 | 31 | 34 | 30 |
| No. of intravitreous melphalan hydrochloride injections | 577 | 622 | 672 | 353 | 169 | 167 | 198 | 162 | 123 | 158 |
| No. of intravitreous topotecan hydrochloride injections | 72 | 0 | 97 | 121 | 45 | 0 | 0 | 0 | 0 | 0 |
| No. of intravitreous methotrexate sodium injections | 0 | 0 | 0 | 17 | 0 | 0 | 0 | 0 | 0 | 0 |
| No. of extraocular events at needle site | 0 | 0 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
| Gauge of needle | 33 | 33 | 30 | 32 | 32 | 33 | 32 | 30 | 32 | 30 |
| Length of needle, mm | 12.7 | 12 | 12 | 13 | 13 | 9 | 13 | 13 | 12 | 12 |
| Injection site distance to the limbus, mm | 3.0 | 3.0-3.5 | 2.0-3.0 | 3.0 | 3.0 | 2.5-3.5 | 2.5 | 3.5 | 3.25-3.5 | 3.0 |
| Dose of drug injected, μg | ||||||||||
| Melphalan hydrochloride | 20-30 | 20-40 | 30 | 20 | 20-30 | 20-30 | 20-30 | 20 | 20-40 | 30 |
| Topotecan hydrochloride | 20 | NA | 20 | 20 | 20 | NA | NA | NA | NA | NA |
| Methotrexate | NA | NA | NA | 400 | NA | NA | NA | NA | NA | NA |
| Volume range of drug injected, mL | 0.04-0.07 | 0.10-0.20 | 0.08 | 0.1 | 0.05-0.1 | 0.1-0.15 | 0.1 | 0.1 | 0.1-0.125 | 0.05 |
Abbreviation: NA, not applicable.
Table 2 gives additional details of the injection method. Cryotherapy was used at the injection site before removal of the needle at all centers. At 7 centers (70%), the IOP was routinely lowered before injection: at 5 centers (40%), paracentesis was used (at 3 of these, higher drug volumes of 0.1-0.2 mL were used); 2 (20%), ocular massage; and 1 (10%), both. At 3 centers (20%), the IOP was not routinely lowered unless 2 drugs were injected or if the IOP was higher than 20 mm Hg. Chemotherapy was injected into the subconjunctival space at only 1 center (10%). At 60 centers (60%), eyes were routinely irrigated after the injection. At most centers, the same entry site was used for repeated injections (7 centers [70%]), ultrasonographic biomicroscopy was performed in eyes with opaque media or absence of mydriasis (7 centers [70%]), and the injection was administered without the use of an operating microscope (6 centers [60%]). At 4 centers (40%), 2 drugs were injected during the same visit, with 2 different needle puncture sites used at all centers. Of the 4 centers in which an operating microscope was used, reflux through the injection site was observed at none. At 1 center (10%), an eye with tumor in the anterior chamber was observed after intravitreous chemotherapy injections; this eye had previous pars plana vitrectomy and cataract extraction with an open posterior capsule. At all centers, at least 2 presumed precautionary measures (cryotherapy, lowering IOP, subconjunctival chemotherapy, ocular surface irrigation, and use of ultrasonographic biomicroscopy) were used, with 3 or more methods used at 8 (80%).
Table 2. Details of the Injection Method Used at Each Center.
| Details on Routine Injection Method | Center Response, No. (%) (N = 10) |
|
|---|---|---|
| Yes | No | |
| Cryotherapy at injection site | 10 (100) | 0 |
| Routinely lower IOP before injection | 7 (70) | 3 (30)a |
| Lower IOP with paracentesis | 5 (50) | 5 (50) |
| Lower IOP with ocular massage | 3 (30) | 7 (70) |
| Inject drug into subconjunctival space | 1 (10) | 9 (90) |
| Each injection performed at same entry site | 7 (70) | 3 (30) |
| Procedure performed under the operating microscope | 4 (40) | 6 (60) |
| UBM performed in eyes with opaque media or absence of mydriasis | 7 (70) | 3 (30) |
| 2 Drugs ever given during the same visit | 4 (40) | 6 (60) |
Abbreviations: IOP, intraocular pressure; UBM, ultrasonographic biomicroscopy.
Two centers used IOP-lowering techniques if IOP was greater than 20 mm Hg or if injecting 2 drugs.
Discussion
To minimize extravasation of tumor at the injection site, a number of centers introduced precautionary practices to lower the risk of extraocular extension, including injecting in a tumor-free location,6,7 subconjunctival chemotherapy,8 paracentesis or ocular massage to lower IOP,6,7,9 cryotherapy to needle tract,9 copious irrigation of the eye,9 swab to injection site to wipe away vitreous,7 and histoacryl glue (in eyes undergoing biopsy).10 Some of these techniques were re-established by Munier et al11 in their 2012 seminal article and are used currently by centers, including those in the present study12,13 (Table 2). Despite the use of these safety-enhancing techniques, questions still exist regarding extraocular extension with intravitreous injections.
Since the resurgence of intravitreous chemotherapy injections in 2012, there have been few dedicated reports on this subject. The studies that exist are limited by small cohorts and use an indirect method (tumor presence in needle of ocular washings) or are restricted to a literature review.3,14 Two cases of extraocular tumor extension detected early in the largest cohort worldwide, to our knowledge, from Japan have been reported—the first case with histopathologic confirmation15 and the second case with extraocular extension that could not exclude intravitreous injection as a contributing factor.3
The present study more than doubles the published meta-analysis cohort and had no cases of extraocular extension in 3553 injections into 704 eyes of 655 patients. These results come from 4 different continents and 2 World Bank income levels (high and middle income). The study revealed a low risk of extraocular extension with intravitreous chemotherapy in eyes with retinoblastoma. It is unclear which and how many of the precautionary measures contributed to the safety of this technique. Perhaps injecting 2 drugs into an eye involves 2 injections and potentially doubles the risk of a single injection; however, no events of extraocular extension were found in eyes receiving a single or double injection. Finally, it is also unknown whether intravitreous chemotherapy can sterilize the needle tract, and because no events were seen in eyes receiving melphalan alone and topotecan alone, it is unknown whether this possible tract sterilization is associated with the chemotherapy agent used (an alkylating agent vs topoisomerase inhibitor).
Limitations
This study is limited by its retrospective, physician-reported methods. It may have benefited from a prospective approach with central review of each case.
Conclusions
This collaborative study of 10 high-volume retinoblastoma centers using at least 2 presumed safety-enhanced techniques for intravitreous chemotherapy injections revealed a low risk of extraocular extension. According to the rule of 3, this finding suggests that the risk is no greater than 0.08%.4 The use of a safe injection practice, including injecting into a tumor-free location of the eye, is recommended and, in this series, was associated with a low risk of tumor externalization. The specific utility of each precautionary method and confirmation of these findings warrant further evaluation.
References
- 1.Karcioglu ZA. Fine needle aspiration biopsy (FNAB) for retinoblastoma. Retina. 2002;22(6):707-710. [DOI] [PubMed] [Google Scholar]
- 2.Honavar SG, Shields CL, Shields JA, Demirci H, Naduvilath TJ. Intraocular surgery after treatment of retinoblastoma. Arch Ophthalmol. 2001;119(11):1613-1621. [DOI] [PubMed] [Google Scholar]
- 3.Smith SJ, Smith BD. Evaluating the risk of extraocular tumour spread following intravitreal injection therapy for retinoblastoma: a systematic review. Br J Ophthalmol. 2013;97(10):1231-1236. doi: 10.1136/bjophthalmol-2013-303188 [DOI] [PubMed] [Google Scholar]
- 4.Hanley JA, Lippman-Hand A. If nothing goes wrong, is everything all right? interpreting zero numerators. JAMA. 1983;249(13):1743-1745. [PubMed] [Google Scholar]
- 5.Ericson LA, Rosengren BH. Present therapeutic resources in retinoblastoma. Acta Ophthalmol (Copenh). 1961;39:569-576. [DOI] [PubMed] [Google Scholar]
- 6.Suzuki S, Kaneko A. Management of intraocular retinoblastoma and ocular prognosis. Int J Clin Oncol. 2004;9(1):1-6. [DOI] [PubMed] [Google Scholar]
- 7.Kivelä T, Eskelin S, Paloheimo M. Intravitreal methotrexate for retinoblastoma. Ophthalmology. 2011;118(8):1689-1689, 1689.e1-1689.e6. doi: 10.1016/j.ophtha.2011.02.005 [DOI] [PubMed] [Google Scholar]
- 8.Seregard S, Kock E, af Trampe E. Intravitreal chemotherapy for recurrent retinoblastoma in an only eye. Br J Ophthalmol. 1995;79(2):194-195. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Chévez-Barrios P, Chintagumpala M, Mieler W, et al. . Response of retinoblastoma with vitreous tumor seeding to adenovirus-mediated delivery of thymidine kinase followed by ganciclovir. J Clin Oncol. 2005;23(31):7927-7935. doi: 10.1200/JCO.2004.00.1883 [DOI] [PubMed] [Google Scholar]
- 10.Char DH, Miller TR. Fine needle biopsy in retinoblastoma. Am J Ophthalmol. 1984;97(6):686-690. [DOI] [PubMed] [Google Scholar]
- 11.Munier FL, Soliman S, Moulin AP, Gaillard M-C, Balmer A, Beck-Popovic M. Profiling safety of intravitreal injections for retinoblastoma using an anti-reflux procedure and sterilisation of the needle track. Br J Ophthalmol. 2012;96(8):1084-1087. doi: 10.1136/bjophthalmol-2011-301016 [DOI] [PubMed] [Google Scholar]
- 12.Shields CL, Douglass AM, Beggache M, Say EAT, Shields JA. Intravitreous chemotherapy for active vitreous seeding from retinoblastoma: outcomes after 192 consecutive injections: The 2015 Howard Naquin Lecture. Retina. 2016;36(6):1184-1190. doi: 10.1097/IAE.0000000000000903 [DOI] [PubMed] [Google Scholar]
- 13.Francis JH, Xu XL, Gobin YP, Marr BP, Brodie SE, Abramson DH. Death by water: precautionary water submersion for intravitreal injection of retinoblastoma eyes. Open Ophthalmol J. 2014;8:7-11. doi: 10.2174/1874364101408010007 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Francis JH, Mondesire-Crump I, Marr BP, Brodie SE, Abramson DH. Cytopathological evaluation of ocular surface and needle washings following intravitreal melphalan injections for retinoblastoma. J Pediatr Ophthalmol Strabismus. 2016;1–3:1-3. doi: 10.3928/01913913-20160122-04 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Suzuki S, Aihara Y, Fujiwara M, Sano S, Kaneko A. Intravitreal injection of melphalan for intraocular retinoblastoma. Jpn J Ophthalmol. 2015;59(3):164-172. doi: 10.1007/s10384-015-0378-0 [DOI] [PubMed] [Google Scholar]