0 Ophthalmic Genetics 0167-67841941

~ ~ $ 3 . 5 0 radiation therapy Recurrence of unilateral retinoblastoma following

Ophthalmic Genetics - 1994, Vol. 15, No. 3/4pp. 107- rr3 0 Bolus Press Buren (The Netherlands) 1994

Accepted 27 September 1994

David H. Abramson’ Camille A. Servodidiol Anthony R. De Lillol

Lisa S. Gamelll Erik F. Kruger’

Beryl McCormick‘

Department of Ophthalmology, The New York Hospital/Cornell Medical Center

Department of Radiation Oncology, Memorial Hospital Sloan Kettering Cancer Center

New York, NY 10021, USA

Abstract A retrospective analysis of 65 unilateral retinoblastoma patients treated initially with external beam radiation, revealed that 25 eyes (38.5%) developed local recurrence of retinoblastoma. The mean age at diagnosis was 1.8 years for patients who developed recurrences vs. 0.9 years for those who did not. Ninety-six percent of the recurrences occurred less than two years from the age at diagnosis; the amount of time from the end of external beam radiation treatment until a tumor recurred was independent of the age at diagnosis. The initial largest basal diameter was 10.7 DD for tumors which later recurred and 5.9 DD for tumors that were cured. Sixty-nine percent of eyes in groups 11 I - v had tumor recurrence, and 10% of eyes in groups I - I I had recurrence. All but one eye (24 eyes) that developed recurrence were enu- cleated. Family history of retinoblastoma, location of the tumor, gender, and laterality did not significantly correlate with the mean age of initial diagnosis for tumors that recurred or the mean time of onset for recurrence.

Correspondence and reprint requests to: David H. Abramson, M.D., 70 E. 66th St., New York, N.Y. 10021, USA

Key words Cancer; external beam radiation; indirect ophthalmoscopy; radiation; Reese-Ellsworth classification; retina; retinoblastoma; tumor re- currence.

Introduction Bilateral retinoblastoma has been treated with external beam radiation since 1903.l In the early days of its use, tumors responded well but ocular radiation complications were In recent years, re- duction of dose, fractionation, more appropriate portals and choice of inci- dent energy have reduced most of these undesired effect^.^'^ When bilateral retinoblastoma is primarily treated with external beam radiation, some eyes require additional treatment such as a second course of radiotherapy; cryo-

Unilateral retinoblastoma 107

Oph

thal

mic

Gen

et D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

acqu

arie

Uni

vers

ity o

n 10

/28/

14Fo

r pe

rson

al u

se o

nly.

therapy: photocoagulation: brachytherapy'0311 or enucleation. The reasons for these additional treatment modalities include unsatisfactory regression of the irradiated tumor, de n o w tumor growth, or local recurrence of tumor growth after a quiescent period (or non-cure). l2

In recent years, unilateral retinoblastoma has been selectively managed without primary en~c1eation.l~ There were three reasons for trying to treat unilateral retinoblastoma without primary enucleation: (I) The hope that, in carefully selected cases, techniques other than enucleation might prevent loss of the organ, (2) The hope that techniques other than enucleation would allow for preservation of useful vision, and (3) The fear that if the fellow eye were to subsequently develop retinoblastoma that the vision and success rate in the first eye might have been better than that of the second, initially uninvolved eye. Although many techniques have been used to treat unilateral retinoblas- toma without primary enucleation the most commonly employed one has been external beam radiati~n.'~

As a result of widespread experience with radiation in cases of bilateral retinoblastoma, the causes for local recurrence following treatment are known. Two features appear to determine which tumors will recur: location of the treated tumor and size of the t ~ m o r . ~ " ~ The more anterior tumors recur more commonly and larger tumors recur more commonly. Similar informa- tion for unilateral retinoblastoma is not known.

There is little known about the time course for recurrences of unilateral or bilateral retinoblastoma after primary treatment with external beam radia- tion. Schipper'' reported that 23 (59%) of 39 irradiated eyes required addi- tional treatment for persistent or recurrent tumors, and were retreated with- in six months after radiotherapy. All but four eyes were in bilateral patients. Zelter et all6 described seven of 14 irradiated patients (11 bilateral, three uni- lateral) who developed recurrence or tumor persistence ranging from two to 30 months after external beam radiation treatment.

The purpose of this retrospective study was to define which unilateral patients have recurrences following primary external beam radiation; which tumors recurred, and over what time frame these recurrences occur. This in- formation might cause us to modify which patients and which tumors we treat primarily with external beam radiation and might cause us to modify our post treatment timing of follow-up examinations.

Materials and methods A retrospective chart review was performed on all unilateral retinoblastoma patients on file at the Ophthalmic Oncology Center at the New York Hospital/Cornell Medical Center who were initially treated with external beam radiation between 1958 and 1988 (to allow a mini- mum follow-up of five years). No patient was lost to follow-up. All patients received 3,500-4,500 cGy over a 33 to &week period with lateral photons which was supplemented with anterior electrons, as previously During follow-up, patients were examined under general anesthesia using the indirect ophthalmoscope. Recurrences were defined as tumors arising from the original tumor focus including vitreous seeds that subsequently implant- ed and grew. The authors were able to discern local failures from new tumor development with greater than 95% certainty. Data were collected on the age of the patients at diagnosis and recurrence, Reese-Ellsworth (R-E) Group at diagnosis, laterality of the involved eye, gender of the patient, presence or absence of family history of retinoblastoma, and size and location of the ori- ginal tumor.

I08 D. H. Abramson eta[.

Oph

thal

mic

Gen

et D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

acqu

arie

Uni

vers

ity o

n 10

/28/

14Fo

r pe

rson

al u

se o

nly.

Mean Mean Male/ O.D./ Positive No. Primary Age Age at L TD

No. Female O.S. Fam Hx TumorFoci ntDx. Recurr D D.

Recurrence 25 8/17 14/11 z 1.5* 1.8 y 2.8 y** 10.7 No recurrence 40 29/11 26/14 15 1.3 0.9 Y 5.9 Significance N.S. N.S. p<o.o5 p < 0.01 *** p < 0.001

* Among ten patients with more than one primary tumor focus, 6 (60%) developed recurrence. Eighteen (34%) of 53 patients with one primary tumor focus developed recurrence. The relative risk of developing tumor recurrence for a patient with more than one primary focus is 2.9. However, chi-squared analysis revealed no significant difference between these two groups.

No patient who was older than seven years of age at diagnosis recurred after treatment. Under the age of seven there was no correlation between the age at diagnosis and time to recur.

'I;** Ninety-two percent of the patients with recurrence were diagnosed at age 3.5 years or younger (Fig. 2 ) . Ninety-six percent of the recurrences occurred less than two years from diagnosis (Fig. 3).

** The mean onset for tumor recurrence was 1.0 years. All recurrences occurred less than 2.75 years from the time of diagnosis (Fig. I).

Results Of the 65 patients included in the study, 25 (38%) developed local tumor recurrence after irradiation. The characteristics of the patients are presented in Table I and Table 2.

Eight (53%) of 15 anterior tumors recurred and 17 (34%) of 50 posterior tumors failed to regress. The relative risk of anterior: posterior tumor failure was 2.2. However, chi-squared analysis showed no significant relationship between initial tumor location and the tendency to recur after irradiation

T A B L E I. Local tumor recurrence after irradiation. LTD=largest tumor diameter. DD=disc diameter

Number

1

of

Fzg I The time interval until onset of tumor recurrence after initial diagnosis and the percentage of all tumor recurrences at that interval.

patients

Age at initial RB diagnosis (yrs.) Time for recurrence (YrS.) a

Unilateral retinoblastoma 109

Oph

thal

mic

Gen

et D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

acqu

arie

Uni

vers

ity o

n 10

/28/

14Fo

r pe

rson

al u

se o

nly.

Fiz 2. Tumor 1 recurrence related to

20

P l 8 E 16 3 E 1 4

; 10

Y E

% 4

'5

e ' 6 a

* 2

0 -1 - 2 ' " " X c 9 U ) b ~ V ) ~ W ~ b I ; ; time interval after initial diagnosis and - 0

treatment. 2 Age at initial RB diagnosis (yrs.)

(p > O . I in all analyses). The mean basal diameter of tumors located anterior to the equator which recurred was the same as the mean basal diameter of anterior tumors that did not recur (mean = 12 DD). The mean basal dia- meter of tumors located posterior to the equator which recurred, however, was significantly larger than the mean basal diameter of posterior tumors that did not recur (10 DD vs. 5 DD,p <0.05).

Fate of eyes with recurrences Twenty-four of the eyes that developed recur- rence (96%) were eventually enucleated. Only one eye (2.5%) of the 40 that did not recur was enucleated (for development of a new tumor which failed treatment).

Patient survival None of the 65 patients has died of metastatic retinoblasto- ma to date. One patient is alive two years after the development of orbital retinoblastoma after intensive chemotherapy. One patient (without recur- rence) has developed a second, non-ocular malignancy.

The relationship between Reese-Ellsworth Classification and recurrence is presented in Table 3.

100 -

Fig. 3. The percentage of patients developing new tumors when initially diagnosed at a particular age.

5 .- b

3

90

80

70

60

5 0

4 0

30

20

10

0

<2 >2

T i m e for onset of tumor recurrence (yrs.)

I 1 0 D. H. Abramson et al.

Oph

thal

mic

Gen

et D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

acqu

arie

Uni

vers

ity o

n 10

/28/

14Fo

r pe

rson

al u

se o

nly.

T A B L E 2. Locationoftumor recurrences. P E / A E N / T S / I

Recurrences 1718 41 I3 215 No recurrences 3311 5 / 1 2 517

PE: Posterior to equator; AE: anterior to equator N: nasal; T: temporal (using macula as reference point) S: superior; I: inferior

Discussion The majority of unilateral retinoblastoma eyes that we have treated primarily with external beam radiation (62%) did not develop local recurrence^.'^ In a previous paper we emphasized that, using this cautious approach, we did not have any deaths from metastatic retinoblastoma to date. Now, ten years later, with longer follow-up, we have still not had a death from metastatic retinoblastoma in this carefully selected group of patients. The reader is cautioned that more than 90% of cases of unilateral retinoblas- toma we manage are primarily managed with primary enucleation and should not infer that this approach is safe for all patients with unilateral retinoblastoma. The information in this paper has helped us delineate which patients have recurrences, which tumors recur and over what time frame re- currences occur.

The patients whose tumors recurred were older than those who did not recur. The mean age of those who had recurrences was 1.8 years vs. 0.9 years for those who did not have them. Ninety-two percent of the patients with re- currence were diagnosed by 3.5 years and 96% of those with recurrence were diagnosed by the age of two years. Since the average age for our unilateral patients is two years, it is clear that the patients we chose to treat were young- er than our average unilateral patient. On the other hand, there were no recur- rences in unilateral patients who were initially treated after the age of seven years. Unilateral patients who recur are older (but less than seven y.0.) than those who do not recur after radiation.

The tumors that recurred were larger in size than those that did not recur (10.7 DD vs. 5.9 DD). Seventy-five percent of tumors that recurred had an initial basal diameter equal to or greater than 10 DD. When looking at the Reese-Ellsworth Classification of tumors, no Group I tumor recurred. Only 16% of Group I I tumors recurred. In contrast 86% of Group I V and 71% of Group v tumors recurred. In our earlier report we found that 10 of 39 patients with unilateral Group I - I I I tumors required additional treatment; 10 of 14 patients with Group I V - v tumors required additional treatment.

T A B L E 3. Distribution of irradiated tumors and recurrences according to Reese-Ellsworth Group

Number ofpatients Total recurrences %of recurrences

R-EGroup: I 14 0 0

I1 19 3 16 I11 I 1 6 55 IV I 6 86

v I4 I 0 71

Unilateral retinoblastoma I 1 1

Oph

thal

mic

Gen

et D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

acqu

arie

Uni

vers

ity o

n 10

/28/

14Fo

r pe

rson

al u

se o

nly.

- 15 I A,. e

I - -

ln l n - G : e w G : ? m n ? 2 k F F N N

Fig. 4. Time course for the 0 0 development of tumor recurrences. 4 Time for on& of tumor recurrence (yrs.)

These results parallel our findings of primary radiation therapy for patients with bilateral retinobla~toma. '~-~~

There was a suggestion from our data that location and multifocality were also correlated with recurrences. The relative risk for developing tumor recur- rence when theeyecontainedmultiplefociwas 2.9. Similarly the relative risk for developing recurrence was 2.2 when the tumors were anterior to the equator.

The time course for the development of recurrences was revealed by this study. All recurrences occurred less than 2.75 years from diagnosis; 96% occurred less than two years from diagnosis and 72% recurred less than one year from diagnosis. The mean interval from treatment to recurrence was 1.0 years and the mean age at recurrence was 2.8 years. This is consistent with prior studies done on fewer It emphasizes that the clinician is justified in frequent examinations under anesthesia for at least three years after treatment and suggests even more frequent examinations in the first year after treatment.

Although no patient has developed metastatic retinoblastoma to date, 24 (96%) of the eyes that failed initial radiation ultimately came to enucleation. Radiation therapy in patients with the germinal mutation for retinoblastoma carries with it an increased incidence of second non-ocular radiation trig- gered cancers.21 It was of interest that one patient in this series has developed a second non-ocular cancer. Careful attention to the findings of this study may enable us to save eyes, vision and minimize second tumors by carefully selecting which unilateral patients we treat primarily with external beam radiation.

References Front RadTher Oncol 1972; 6: 346. 3 MacFaul PA, Bedford MA. Ocular

complications after therapeutic irradiation. Br J Ophthalmol 1970; 54:

I Hilgartner HL. Report of a case of double glioma treated with x-ray. Texas

z MerriamGR, Szechter A, Focht EEThe effects of ionizing radiation on the eye.

Med J 1903; 18: 322. 237-241. 4 Reese AB, Hyman GA, Tapley NV,

Forrest AF. The treatment of retino-

I12 D. H. Abramson et al.

Oph

thal

mic

Gen

et D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

acqu

arie

Uni

vers

ity o

n 10

/28/

14Fo

r pe

rson

al u

se o

nly.

blastoma by x-ray and triethylene melanine. Arch Ophthalmol 1958; 60: 897-906.

5 Abramson DH, Jereb B, Ellsworth RM. External beam radiation of retinoblastoma. Bull NY Acad Med 1981; 57: 787-803.

6 Ellsworth RM. Retinoblastoma. Mod Probl Ophthalmol 1977; 18: 94-100.

7 Abramson DH, Ellsworth RM, Rosen- blatt Met al. Retreatment of retino- blastoma with external beam irradia- tion. Arch Ophthalmol1982; 100: 1257- 1260.

Rozakis GW. Cryotherapy for retino- blastoma. Arch Ophthalmol 1982; 100:

9 Abramson DH. The focal treatment of

8 Abramson DH, Ellsworth RM,

1253-1256.

retinoblastoma with emphasis on xenon arc photocoagulation. Acta Ophthalmol 1989; 67: I - 63.

10 Ellsworth RM. The treatment of retinoblastoma. Trans Aust Coll Ophthalmol 1971; 3: 100-110.

11 Abramson DH, Ellsworth RM, Haik BG. Cobalt plaques in advanced retinoblastoma. Retina 1983; 3: 12- 15.

12 Ehlers N, Kaae S. Effects of ionizing radiation on retinoblastoma and on the normal ocular fundus in infants. Acta Ophthalmol 1987; 65: Suppl 181: I - 84.

Ellsworth RM et al. The management of unilateral retinoblastoma without primary enucleation. Arch Ophthalmol

13 Abramson DH, Fusco Marks R,

1982; 100: 1249-1252.

14 Abramson DH. Treatment of retino- blastoma. In: Blodi FC, editor. Retino- blastoma (Contemporary Issues in Ophthalmology). New York: Churchill Livingstone, 1985; 69-93.

15 Schipper J. Retinoblastoma. A medical and experimental study. Thesis. State University of Utrecht 1980.

16 Zelter M, Gonzalez G, Schwartz Let al. Treatment of retinoblastoma: results obtained from a prospective study of 51 patients. Cancer 1988; 61: 153-160.

17 Reese AB, Ellsworth RM. The evalua- tion and current concept of retino- blastoma therapy. Trans Am Acad Ophthalmol Otolaryngol1963; 67: 164 - 172.

18 Abramson DH, Ellsworth RM, Tretter P et al. Treatment of bilateral groups I through 111 retinoblastoma with bilateral radiation. Arch Ophthalmol 1981; 99: 1761-1762.

P et al. Simultaneous bilateral radiation for advanced bilateral retinoblastoma. Arch Ophthalmol 1981; 99: 1763-1766.

Diagnosis and Management of Intra- ocular Tumors. St Louis: CV Mosby,

21 Abramson DH, Ellsworth RM, Tung D et al. Second tumors in retinoblastoma survivors: Are they radiation induced? Ophthalmology 1984; 91: I351 - 1355.

19 Abramson DH, Ellsworth RM,Tretter

20 Shields, JA. Retinoblastoma. In:

1983; 437- 496.

Unilateral retinoblastoma 113

Oph

thal

mic

Gen

et D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

acqu

arie

Uni

vers

ity o

n 10

/28/

14Fo

r pe

rson

al u

se o

nly.