BRENDA GALLIE医生，来自Departments of Ophthalmology & Vision Science, Hospital for Sick Children and University of Toronto, Toronto, Canada. Techna Institute, University Health Network, Toronto, Canada
Pars Plana Vitrectomy and Endoresection of Refractory Intraocular Retinoblastoma
Unplanned intraocular surgery in eyes with active retinoblastoma carries a risk of extraocular extension and metastases. Planned intravitreal chemotherapy (IVitC) has signiﬁcantly revolutionized the treatment of the most intractable feature of recurrent retino-blastoma, vitreous seeds, without evidence of extraocular spread of tumor. One child with recurrent refractory vitreous seeding was treated with pars plana vitrectomy (PPV) with good vision and no active tumor at 26 months of follow-up.
We now report results of planned PPV and endoresection of active retinoblastoma refractory to standard treatments (systemic chemo-therapy, intra-arterial chemotherapy, focal therapy including brachy-therapy, and intravitreal melphalan) in 21 children with only one remaining eye, as an alternative to enucleation (Fig 1). Institutional ethics review board approval was obtained from Beijing Children’s Hospital and Quanzhou Children’s Hospital for this retrospective review of the medical records of children treated with PPV. A team approach included the retinoblastoma expert (J.Z.) who indicated the intended target for PPV and expert retinal surgeons (Q.L., S.W.) who performed the PPV. The children were referred when retinoblastoma treated in multiple institutions was unresponsive to standard therapies. All children had no tumor obstructing visibility of the optic nerve or suggestion of nerve invasion on neuroimaging and no evidence of extraocular disease at any time (before or after PPV).
Before PPV, all children received a median of 6.5 cycles (range, 2-15 cycles) of systemic chemotherapy with carboplatin, etopo-side or teniposide, and vincristine (and cyclophosphamide for patients 5 and 14); 4 eyes experienced retinal detachment (patients 3, 15, 16, and 17); 2 eyes (patients 11 and 21) received IVitC; 1 eye (patient 5) underwent brachytherapy with cataract extraction; and 3 eyes (patients 1, 11, and 20) received intra-arterial chemotherapy.
Three-port PPV was performed with a 23- or 25-gauge cutter probe with a nonvalved cannula with 5 mg/ml melphalan in irrigation ﬂuid. A laser barrier around solid tumors and silicone oil stabilized the retina. To reach anterior segment seeding (patient 4), the lens was removed and the tumor was aspirated from the anterior chamber. Solid active tumor was removed by endoresection. After PPV, 0.2 ml melphalan (5 mg, 25 mg/ml) was injected subconjunctivally at surgical entry points and repeated at the subsequent 2-4 examinations under anesthesia. Intravitreal chemotherapy (melphalan, 20 mg/0.05 ml) was given monthly if extensive tumor had been removed (patients 8-17). Regular examinations under anesthesia followed PPV.
Up to the end of 2016, follow-up time from diagnosis was a median of 5.1 years (range, 2.8-10.7 years); that from PPV was 3.3 years (1.6-4.3 years; Fig 1; Fig S1). One patient (patient 21) was lost to follow-up with recurrent intra-ocular tumor. As far as is known at this time, no patient demonstrated metastases or died. Two eyes (patients 19 and 20) underwent enucleation for intraocular tumor recurrence; 18 eyes were salvaged. Available wide-ﬁeld retinal images of all 21 eyes before PPV and at last follow-up are shown in Figure S2.
The last follow-up visual acuity was functional in 14 of the 18 assessable salvaged eyes (78%; 11 had better than 20/80 vision; 2 had 20/100 vision; 1 had 20/200 vision) and poor in 4 children (3 had counting ﬁngers vision; 1 had light perception vision) (Fig S1).
After PPV, 5 patients required no more treatment (Fig 1A). Two patients (patients 4 and 13) received 1 cycle of systemic chemotherapy because of suspected choroidal involvement at the resection site, and 2 because the other eye had high risk pathology after enucleation (Fig 1B). One patient underwent surgery for retinal detachment (patient 17) and 2 patients underwent cataract surgery (patients 15 and 19). Ten patients with a high burden of vitreous seeds at PPV received adjuvant IVitC 1-4 times (Fig 1C). One patient (patient 18) received IVitC to treat residual vitreous seeds, but still required enucleation. Four patients (patients 12, 13, 16, and 17) received IVitC while silicone oil was in the eye; no retinal toxicity was observed.
The retinoblastoma was not controlled by 1 PPV in 4 patients, all of whom demonstrated recurrence with difﬁcult-to-visualize anterior tumor (Fig 1D). Two patients underwent enucleation for recurrent disease; no tumor was observed in the needle tracts on pathologic examination (patients 18 and 19). Two patients received 1 (patient 20) or 2 (patient 21) additional PPVs for recurrent tumor. Patient 20 achieved 20/100 vision after a second PPV. The parents of patient 21 declined enucleation for recurrence after 3 PPVs, and the child is now lost to follow-up.
Last follow-up vision (Fig 1; Fig S1) was functional in 14 of 18 assessable eyes (78%; 11 with better than 20/80 vision; 2 with 20/100 vision; and 1 with 20/200 vision) and poor in 4 eyes (3 with counting ﬁngers vision; 1 with light perception vision). We observed no dissemination of cancer cells outside the eye at 3.3 years median follow-up after PPV.
The lost-to-follow-up patient (patient 21) received 9 cycles of systemic chemotherapy, enucleation of the other eye, 4 IVitC injections, and 3 PPVs for anterior recurrent tumor that failed to control the cancer (Fig 1D; Fig S2). Enucleation was proposed repeatedly, but declined, and the child is lost to follow-up with active intraocular disease. Unless treatment is ongoing elsewhere, this child may eventually die of recurrent retinoblastoma. The lost-to-follow-up rate in China (2006-2008) was 40% at 5 years after diagnosis, currently estimated at 20% (J. Zhao, unpublished data, 2016), higher than that of our present study (1/21 [5%]).
Organ salvage cancer therapy is common for most cancers, but the vastly different survival rates for intraocular and extraocular retinoblastoma have dissuaded ophthalmologists from exploring so-called lumpectomy for retinoblastoma. In China from 2006 through 2009, 55% of children demonstrated advanced Interna-tional Intraocular Retinoblastoma Classiﬁcation (IIRC) Group E retinoblastoma (enucleation recommended) and 33% demonstrated Group C/D disease (considered safe to attempt salvage). Despite systemic chemotherapy, focal intra-arterial chemotherapy, or both, followed by focal therapy (laser, cryotherapy, and IVitC), recurrence commonly leads to extensive ongoing treatments focused on saving the eye. This large investment may cause parents to resist enucleation when that becomes the safest option.
The careful approach developed by Munier et al for IVitC has no reported occurrence of extraocular spread of tumor. For the many children in China with recurrent disease after all standard therapies, a direct and deﬁnitive therapy to save remaining eyes was needed: PPV and tumor resection achieved this goal for 18 of 21 children with no instance of extraocular spread of tumor. This approach to retinoblastoma depends on multidisciplinary teamwork and careful technique to avoid tumor spread.
Figure 1. DePICTRB (Disease-speciﬁc, electronic Patient Illustrated Clinical Timeline retinblastoma) summaries of children treated with pars plana vit-rectomy (PPV). A, For 5 patients after PPV, no further treatments were required. B, Patients 6 and 7 were treated with systemic chemotherapy with carboplatin, etoposide, and vincristine after PPV because of high-risk pathologic features in the enucleated eye. C, Ten patients were treated with intra-vitreal chemotherapy (IVitC) with melphalan after PPV. D, Four eyes failed primary PPV: patients 18 and 19 underwent enucleation, a second PPV saved patient 20 with 20/50 vision, and patient 21 remains lost-to-follow-up with 20/100 vision after 2 subsequent PPVs. The follow-up from diagnosis and from PPV is displayed by the years between staging examination under anesthesia (EUA) and PPV and the right green line indicating last follow-up. Cancer staging by International Intraocular Retinoblastoma Classiﬁcation (IIRC) and the eighth edition of the American Joint Committee on Cancer Staging Manual.5 Four patients (patients 4, 5, 14, and 21) had a family history of retinoblastoma. CF ¼ counting ﬁngers; NLP ¼ no light perception.
The authors thank Ivana Ristevski for data analysis and ﬁgure development; Cynthia Vandenhoven, Raumil Patel, and Kaitlyn Hougham for ﬁgure development; and Sameh Soliman for manu-script review.
1.JUNYANG ZHAO, MD
2.QIYAN LI, MD
3.SONGYI WU, MD
4.LIWEN JIN, MD
5.XIAOLI MA, MD
6.MEI JIN, MD
7.YIZHUO WANG, MD
8.BRENDA GALLIE, MD
1. Department of Ophthalmology, Beijing Children’s Hospital, Capital Medical University, Beijing, China.
2．Department of Ophthalmology, Beijing Tongren Hospital, Capital Medical University, Beijing, China.
3．Department of Ophthalmology, Quanzhou Children’s Hospital, Quanzhou, China.
4．Department of Ophthalmology, Quanzhou Children’s Hospital, Quanzhou, China.
5．Hematology and Oncology Centre, Beijing Children’s Hospital, Capital Medical University, Beijing, China.
6．Hematology and Oncology Centre, Beijing Children’s Hospital, Capital Medical University, Beijing, China.
7. Department of Paediatrics, Beijing Tongren Hospital, Capital Medical University, Beijing, China.
8. Departments of Ophthalmology & Vision Science, Hospital for Sick Children and University of Toronto, Toronto, Canada. Techna Institute, University Health Network, Toronto, Canada
Both authors contributed equally as ﬁrst authors.
Financial Disclosure(s): The author(s) have no proprietary or com-mercial interest in any materials discussed in this article.
HUMAN SUBJECTS: Human subjects were included in this study. The ethical review board approved the study, with consideration to the tenets outlined in the Declaration of Helsinki.
Conception and design: Zhao, Gallie
Analysis and interpretation: Zhao, Gallie
Data collection: Zhao, Li, Wu, L. Jin, Ma, M. Jin, Wang, Gallie
Obtained funding: none
Overall responsibility: Zhao, Gallie
Brenda Gallie, MD, Department of Ophthalmology and Vision Science, Hospital for Sick Children, 555 University Avenue, Toronto, Ontario M5G1X8, Canada. E-mail: email@example.com
1. Munier FL, Soliman S, Moulin AP, et al. 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.
2. Francis JH, Brodie SE, Marr B, et al. Efﬁcacy and toxicity of intravitreous chemotherapy for retinoblastoma: four-year expe-rience. Ophthalmology. 2017;124(4):488-495.
3. Ji X-D, Lu S-L, Zhao P-Q. Vitrectomy for localized vitreous seeds of retinoblastoma in an only eye. Chinese Med J. 2013;126:2589-2590.
4. Zhao J, Li S, Shi J, Wang N. Clinical presentation and group classiﬁcation of newly diagnosed intraocular retinoblastoma in China. Br J Ophthalmol. 2011;95(10):1372-1375.
5. Mallipatna A, Gallie BL, Chévez-Barrios P, et al. Retino-blastoma. In: Amin MB, Edge SB, Greene FL, eds. AJCC Cancer Staging Manual. 8th ed. New York: Springer; 2017: 819-831.