The Prognostic Value of Testicular Biopsy in Childhood Acute
Lymphoblastic Leukemia: A Report From the Childrens Cancer Study Group By Denis R. Miller, Sanford L. Leikin, Vincent C. Albo, Nigel F. Palmer, Harland N. Sather, and G. Denman Hammond One of the objectives of Childrens Cancer Study Group (CCSG) study 141 (CCG-141) was to determine the frequency of occult testicular leukemia (TL) after 3
years of disease-free survival (DFS) and to retreat boys with occult TL to prolong their subsequent DFS. Of the 494 boys entered on study, 255 (51.6%) were in complete continuous remission (CCR) 3 years after entering remission and an additional eight were in CCR 3 years after localized extramedullary relapse and retreatment; 263 boys were eligible for testicular biopsy. Elective testicular biopsy was performed on 235 (89.4%) boys. Of the 204 (86.8%) boys with negative biopsies, 175 (85.8%) remained in CCR 10 to 12 years after diagnosis and 25 (12.3%) relapsed, 11 (44%)of whom died. Isolated overt TL occurred in four (2.0%) and all remained in CCR 22+ to 60+ months after re-treatment. Of the 26 boys with occult TL, 16 (62%) remained in CCR. Ten (38%) relapsed despite local testicular irradiation and systemic re-treatment;
C
URRENTLY, more than 50% of effectively treated children with acute lymphoblastic leukemia (ALL) are in continuous complete remission (CCR) 3 years or more after diagnosis. The value of "front-end" factors in determining the long-term prognosis of previously untreated children with ALL has attained universal acceptance and use. 1'2 However, until now it has not been possible to identify prognostic factors at the end of scheduled treatment that predict late relapse. Cessation of therapy after 3 years of CCR carries a risk of relapse of 10% to 20%, higher in males than in females."5 The reported rate of occult testicular leukemia (TL) after 2 to 3 years of CCR or overt TL relapse after therapy is discontinued is approximately 10%, and has been reported as high as 40% and as low as 0 . 9 %.6-21 It is difficult to predict which males are at risk of developing TL after cessation of treatment because, except for sex, other reported prognostic factors lose their ability to predict relapse in all sites after 2 years of CCR. 22 It is not certain whether occult TL is a sign of localized or widespread residual disease.12,19 The
six of the 10 died. Of the 26 boys who did not undergo biopsy, 21 (80.8%) remained in CCR; two (7.7%) developed isolated overt TL. DFS after testicular biopsy was significantly better in boys without occult TL (P = .001). Occult TL after 3 years of CCR represents aggressive minimal-residual disease and carries a worse prognosis than absence of TL. Initial treatment should be directed at obviating occult and overt testicular relapse. Conventional therapy as used in this study was suboptimal in preventing subsequent bone marrow (BM) relapse and death. If occult TR is identified during or at the end of planned therapy, a higher salvage rate may require intensified alternate therapy. As such, testicular biopsies may be clinically useful. Further investigation is limited by the relative rarity of, and the lack of identifying features in boys with occult TL. J Clin Oncol 8:57-66. @ 1990 by American Society of ClinicalOncology.
prognostic significance of isolated occult TL while on chemotherapy or at end-therapy has not been determined and continues to be the subject of controversy. 9-2 The clinical use of elective end-therapy testicular biopsies has been questioned by citing a high false-negative rate,2 3 a low incidence of occult TL,19, 20 and the success of treating isolated overt testicular relapse after therapy is discontinued.2 3 In a previously reported Childrens Cancer Study Group (CCSG) study (CCG-101/143), From the Lutheran General Children's Medical Center, Park Ridge, IL; Children's Hospital National Medical Center, Washington, DC; Children'sHospitalofPittsburgh, PA; Children'sHospitalof Columbus, OH; and the University of Southern CaliforniaSchool of Medicine, Los Angeles, CA. Submitted August 9, 1988; accepted August 25, 1989. Supported in part by the Division of Cancer Treatment, National Cancer Institute, National Institutes of Health, Department of Health and Human Services, Bethesda, MD. Address reprint requests to the Childrens Cancer Study Group, 199 North Lake Ave, Third Floor, Pasadena, CA 91101. © 1990 by American Society of Clinical Oncology. 0732-183X/90/0801-0008$3.00/0
Journalof Clinical Oncology, Vol 8, No 1 (January), 1990: pp 57-66
Downloaded from ascopubs.org by RB DRAUGHTON LIBRARY on April 26, 2019 from 131.204.073.184 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
57
58
MILLER ET AL
testicular biopsies were not performed before the cessation of therapy after either 3 years or 5 years of CCR. The overt testicular relapse rate off-therapy was 9.5%, 44% of whom died despite local and systemic retrieval therapy. In this study, CCG-141, open-wedge testicular biopsies were mandatory in all boys remaining in CCR for 3 years and were performed before the randomization to stop chemotherapy or continue chemotherapy for an additional 2 years. Objectives of the study were to determine the frequency of occult TL after 3 years of CCR, and to prolong subsequent recurrence-free survival in boys with occult TL by identifying and treating residual extramedullary disease. PATIENTS AND METHODS CCG-141 was opened in February 1975 and was closed to patient entry in February 1977. The details of the treatment regimens are published' and summarized in Fig 1. Briefly, patients less than 21 years of age were stratified according to low (< 20 x 109/1) or high (a 20 x 109/1) initial WBC count. Children with low initial WBC counts received regimen 1: induction with vincristine, prednisone, and asparaginase, CNS prophylaxis with 24 Gy cranial radiation and weekly intrathecal methotrexate for 6 weeks, and maintenance therapy for the first year with daily mercaptopurine, weekly methotrexate, and monthly pulses of vincristine and prednisone. Patients with high initial WBC counts were randomized to receive either regimen 1 or regimen 2, a more intensive program of induction, CNS prophylaxis, and maintenance Phoas I INDUCTION Day Io 42
"Low" REG I
therapy. After the first year of CCR all patients were maintained on daily mercaptopurine and weekly methotrexate. After 3 years of initial CCR or subsequent (secondary) CCR after localized extramedullary (CNS, testicular, skin, other) relapse, all males were to undergo bilateral openwedge biopsy. Biopsy material evaluated by each participating institution's pathologist was centrally reviewed by the CCSG Pathology Center (CPC) in Columbus, OH. Eligibility for the duration of therapy phase of CCG-141 required that patients have a negative biopsy as interpreted by the member institution and/or the CPC. A fully evaluable patient with a positive biopsy required protocol-prescribed local therapy and systemic reinduction and maintenance. Disposition and treatment were based upon the institution review in over 90% of the cases with positive biopsies. Boys with occult TL received local testicular radiation (20 to 24 Gy), reinduction therapy with vincristine, prednisone, and asparaginase, weekly intrathecal methotrexate for 6 weeks, and daily mercaptopurine, and weekly methotrexate for 3 years or until relapse. Patients with negative biopsies were eligible to be randomized to one of three regimens designed to determine the optimal duration of therapy. In regimen A, all chemotherapy was discontinued; in regimen B, patients received a 4-week course of reinduction therapy (vincristine, prednisone, and asparaginase) and then therapy was discontinued; in regimen C, maintenance therapy was continued for 2 more years and then stopped. Boys with equivocal biopsies were also randomized to regimens A, B, and C. Informed consent was obtained according to the investigator's institutional review board requirements. Comparisons between groups were made using the MantelPeto-Cox summary X2 (log-rank) method of life tableanalysis.24 The standard X2 statistic was used where appropri-
Phase II
Ph.ae III
Phase IV
Phot. V
CNS fx to 2 or onths
MAINTENANCE I to 12 months
MAINTENANCE II to 36 months from corpletiown of CNM thpy
MAINTENANCE III to 60 months from completion of CN th.Mry
PON
Phose VI OBSERVATION
VCR
"LOW GRONP" wBC 2o.0om And no Mediostinal wid*nin)
"Hig
"HIGH GROLPX
WaBC 2000 ond/o, Medilosinal
"w-dninq
M3 doa N · Polients _1
-m-,
OFF STUDY - 2001h
Fig 1.
Study schema, CCG-141.
Downloaded from ascopubs.org by RB DRAUGHTON LIBRARY on April 26, 2019 from 131.204.073.184 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
59
TESTICULAR BIOPSY IN CHILDHOOD ALL ate. The end points evaluated in tlhe life-table analysis included the following: (1) disease-fr ee survival (DFS) or time to first occurrence of bone marro' wrene of bone w, CNS, testicular, or testicular, row other extramedullary relapse; (2) occur rence of bone marrow (BM) relapse (> 25% blasts); (3) o,ccurrence of isolated symptomatic CNS relapse; (4) occurr ence of isolated overt testicular relapse; and (5) death.
tient permission for testicular biopsy was refused in 26 boys (9.9%) who then were not eligible for randomization. All boys who refused biopsy were
The data represent patient ffollow-up through June 1987. All patients are no,w 10 to 12 years postdiagnosis, and 3 to 9 years off-therapy. The median follow-up is 60+ mont ths from the time of testicular biopsy. Of the 494 boys entered on st:udy, 465 (94.1%) achieved a complete remission and 255 (51.6%) were in CCR 3 years after completing CNS therapy. Eight additional boys who had isolated extramedullary relapses during primary maintenance therapy (CNS, five; tes ticular, two; gingiva/palate, one) received prrotocol-prescribed therapy for an additional 36 ± 3 months and were eligible to undergo biopsy (Table 1). Open-wedge testicular bio psies were performed in 235 (89.4%) of the 2( 63 eligible boys, of whom 196 (83.4%) had bilater;al biopsies and 14 (6.0%) had unilateral biopsies;. Information regarding unilaterality or bilate :rality of biopsies was unavailable in 25 (10.6%) ).Parental or pa-
followed on the duration of therapy study. Of all the males biopsied, 204 (86.8%) had no testicular involvement and 31 (13.2%) had occult TL. Of the 230 males enrolled in the duration of therapy study, 137 (59.6%) were randomly assigned to one of the three study regimens; of the 93 boys (40.4%) who were assigned nonrandomly, 67 had negative biopsies, 24 were not biopsied, and two had discordant positive biopsies. The nonrandom assignment rate of 40% is in accord with other CCSG clinical trials and reflects primarily parental reference, and also was in accord with informed consent. The DFS rates in boys with negative testicular biopsies who were assigned randomly (86.7%, observed: expected ratio, 0.83) and nonrandomly (85.6% observed: expected rate, 0.88) were virtually identical. Thus, all boys with negative biopsies were analyzed together. Thirty-one boys had isolated occult TL (13.2% of those biopsied). Five boys with occult TL were excluded from further analysis, three with discordant results and two with biopsies performed before or after the protocol-prescribed time point. The protocol stipulated that biopsies be per-
Table 1. Testicular Biopsy Study: Summary Data
formed within 3 to 5 months of the predeter-
RESULTS
Patients
rNumber
Entered on study CCG-141 Completed induction with CR In CCR at 36 -- 6 months after CNS therapy In secondary CCR 36 ± 6 months after isolated extramedullary relapse* Eligible for testicular biopsy No biopsy performed Biopsied Negative biopsy Positive biopsy Protocol compliant Protocol deviations Entered on duration of therapy study Randomized Nonrandomized
494
Percent\
100
465
94.1
255
51.6
8*
1.6*
263
53.2
26 235 204 31
9.9 89.4
26
11.1
86.8
13.2
230
mined completion date of primary (3 months) or A
+
LL""'" Y 1111111111 ocs naJ---nh)maneaneter
lllnllllllnlll.l
Tt.h
Illnllv
II
biopsy was performed too early in one case (29
months) and too late (44 months) in another. Three patients with discordant biopsy (institu-
tion negative-CPC positive, 2; institution positiveCPC negative, 1) were entered on the duration of therapy study and had therapy discontinued, one randomly and two nonrandomly. Thus, there were 26 boys with isolated occult TL treated per protocol and fully evaluated for fate after biopsy. Two boys with occult TL were treated with orchidectomy of the involved testis, systemic reinduction, and maintenance therapy and were radiation was included in the analysis. Testicular
137
59.6
not given.
93
40.4
The clinical characteristics of the 31 bovs with isolated occult TL, 204 boys with negative biopsies, and the 26 who were not biopsied are j
Abbreviation: CR, complete response. *Isolated extramedullary relapses: CNIS,five; overt testicular, two; gum/palate ulcer, one; (two of the eigght had occult ThL). biopsy (29 months), one; late biopsy (44 months), one; discordant review, entered on duration of therapy study, three
"tEorly
(therapy discontinued).
.......
compared in Table 2. Only one boy with occult TL had an initial WBC count greater than 50 x
109/1 and none had features of the leukemia/
Downloaded from ascopubs.org by RB DRAUGHTON LIBRARY on April 26, 2019 from 131.204.073.184 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
60
MILLER ET AL Table 2. Clinical Characteristics of 235 Boys Undergoing Testicular Biopsy After 3 Years CCR and 26 Boys Not Biopsied
Characteristics Initial WBC count - 20 x 10'/L > 20 x 109/L Age < 12 months 13 to 120 months > 120 months Prognostic group Good Average Poor FAB classification L1 L2 Unknown IG level Normal Decreased Unknown Day-14 BM M1 M2/M3 Unknown Hemoglobin < 10 gm/dL - 10 gm/dL Splenomegaly, marked/moderate Mediastinal mass, present Leukemia/lymphoma syndrome
Occult Testicular
Negative Biopsy
Leukemia (N = 31)*
(N = 204)
No Biopsy (N = 26)
N
%
N
%
N
5
26 5
83.9 16.1
153 51
75.0 25.0
20 6
76.9 23.1
0 28 3
0.0 90.3 9.7
4 166 34
2.0 81.4 16.6
0 22 4
0.0 84.6 15.4
11 19 1
35.5 61.3 3.2
55 125 24
27.0 61.3 11.7
10 12 4
38.5 46.2 15.4
27 2 2
87.0 6.5 6.5
163 24 17
79.9 11.8 8.3
21 3 2
80.8 11.5 7.7
21 4 6
67.7 12.9 19.4
129 45 30
63.2 22.1 14.7
13 3 10
50.0 11.5 38.5
18 6 7
58.1 19.3 22.6
120 53 31
58.8 26.0 15.2
17 4 5
65.4 15.4 19.2
28 3 13 0 0
90.3 9.7 41.9 0.0 0.0
167 37 87 11 15
81.9 18.1 42.6 5.3 7.3
22 4 9 1 1
84.6 15.4 34.6 3.8 3.8
Abbreviations: CCR, complete continuous remission; FAB, French-American-British; IG, immunoglobin. *Includes five ineligible boys, three with discordant positive biopsies entered on duration of therapy study, one with early biopsy (29 months), one with late biopsy (44 months). "fPrognostic group definitions: good, initial WBC count < 10,000//uL, age 3 to 7; poor, initial WBC count n 50,000/tL; average, all other patients. tSee text for definition.
lymphoma syndrome as defined in this study1 (presence of three of the five following features: splenomegaly below the umbilicus, marked lymphodenopathy [visible, 3 cm single node or 5 cm mass], mediastinal mass, hemoglobin _ 10 gm/ dl, or WBC count > 50 x 109/1). The distribution by initial WBC count, age, prognostic group, lymphoblast morphology, immunoglobulin levels, results of the day-14 BM, hemoglobin level, degree of splenomegaly and lymphadenopathy, presence of mediastinal mass, or presence of leukemia/lymphoma syndrome was not significantly different in boys with occult TL compared with those with negative biopsies or those not biopsied. Cell surface marker and chromosomal analyses were not required in CCG-141. Correla-
tions between immunologic phenotype, karyotype, and occult TL were not determined. ConcordanceStudy Of the 235 testicular biopsies performed, 172 (72.6%) were reviewed by the CPC, of which 163 (94.7%) were assessable (Table 3). The review was concordant with the member institution in 150 biopsies (92.0%) and discordant in 13 (8.0%). Unilateral rather than bilateral involvement was noted in only two of 144 (1.4%) bilateral biopsies reviewed by the CPC. Of the 31 boys with isolated occult TL, five biopsy samples that were positive at the member institution were not reviewed centrally. Of the remaining 26 biopsies, 13 biopsies were fully concordant, four were
Downloaded from ascopubs.org by RB DRAUGHTON LIBRARY on April 26, 2019 from 131.204.073.184 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
61
TESTICULAR BIOPSY IN CHILDHOOD ALL Table 3. Testicular Biopsy: Concordance Study Number Males biopsied
235
Percent 100
Not reviewed/inevaluable by CPC Institution +t Institution -t Assessable by CPC Concordant review Institution +/CPC Institution -/CPC Discordant review Institution +/CPC Institution -/CPC
+ +
74* 7 67 163 150 13 137 13 4 9
31.2 2.9 28.3 68.8 92.0t (63.8)§ (5.5) (58.3) 8.0f (5.5) (1.7) (3.8)
*Includes nine specimens reviewed by CPC, but inassessable. t +, positive; -, negative. *Percent of assessable specimens. §Number in parentheses equals percent of total biopsy specimens.
interpreted as positive at the member institution and negative at the CPC, and nine were considered negative at the member institution and positive by the CPC. Fate After TesticularBiopsy Of the 204 boys with negative testicular biopsies (Table 4), 175 (85.8%) remained in CCR, 25 (12.3%) relapsed, and four (2.0%) died of nonleukemic causes (three with malignant brain tumors, one of disseminated varicella). Isolated BM relapse occurred in 13 boys, 1 to 36 months (median, 11 months), following testicular biopsy and nine have died. Concomitant BM and overt TL occurred in four boys, nine to 47 months after testicular biopsy, and one has died. Isolated overt TL occurred in four boys with prior negative biopsies, a "false-negative" rate of 2.0%. The Table 4. Fate After Negative Biopsy Number Negative CCR Relapse BM BM/overt testicular Isolated overt testicular Isolated CNS Alive after relapse Dead after relapse Death unrelated to leukemia
204 175 25 13 4 4 4t 14 11 4*t
Percent 200 85.8 12.3 6.4 2.0 2.0 2.0 56.0 44.0 2.0
*Brain tumors, three; disseminated varicella, one. tlncludes one patient with CNS relapse 16 months after testicular biopsy, death due to high-grade astrocytoma 25 months after biopsy.
overt testicular relapses occurred from 14 to 49 months after testicular biopsy, but all four patients remain in CCR 22+ to 60+ months after protocol-prescribed treatment. Isolated CNS relapses occurred in four boys, 3 to 56 months after biopsy. Three remain in CCR after retreatment. One, who developed a right fronto-parietal lesion diagnosed by needle biopsy as "CNS leukemia" 16 months after testicular biopsy, died of a high-grade astrocytoma 9 months later. The total relapse rate was significantly higher in boys with negative testicular biopsies who had chemotherapy randomly or nonrandomly discontinued after 3 years (regimen A, 13 of 84, 15.5%) compared with boys who were treated for 5 years (regimen C, four of 64, 5.8%, P < .05). The difference was not significant when only randomly-assigned patients were compared. The fate of boys with isolated occult TL is shown in Table 5. Of the 26 boys with occult TL, 16 (61.5%) remained in CCR after local radiation (or orchidectomy in two), systemic reinduction, CNS prophylaxis with intrathecal methotrexate, and 2 to 3 years of maintenance therapy. The median duration of follow-up after testicular biopsy was 60+ months (range, 17 to 73+ months). Ten (38.5%) patients with occult TL relapsed: nine sustained isolated BM relapses 12 to 30 months (median, 17 months) after occult TL, six of whom have died. One patient developed an isolated CNS relapse 23 months after detection of occult TL and remains in complete remission 41 + months. At the time of the latest update all boys who had occult TL and were re-treated were off therapy. The relapse rate did not differ in patients in whom the determination of a positive biopsy was made by the member institution or by the central review pathologist; relapses were evenly distributed between institution-positive and CPC-positive biopsies. Of the 26 patients who did not have testicular Table 5. Fate After Positive Testicular Biopsy and Re-treatment
Positive biopsy CCR Relapse BM CNS Dead after relapse
Number
Percent
26 16 10 9 1 6
100 61.5 38.5 34.6 3.9 60
Downloaded from ascopubs.org by RB DRAUGHTON LIBRARY on April 26, 2019 from 131.204.073.184 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
62
MILLER ET AL
Table 6. Fate After No Biopsy (Refusal/Protocol Violation) Number
Percent
26 21 5 2 2 0 1 1 4
100 80.8 19.2 7.7 7.7
No biopsy CCR Relapse Isolated BM Isolated testicular Isolated CNS BM plus testicular Dead after relapse Alive after relapse
3.8 20.0 80.0
biopsies 21 (80.8%) remained in CCR 22+ to 84+ months (median, 57 months) from the completion date of 3 years of maintenance therapy (Table 6). Five patients (19.2%) relapsed, in three of whom therapy was nonrandomly discontinued and in two of whom maintenance therapy was continued for an additional 2 years. Four of the five relapses occurred after therapy had been discontinued. The only death occurred in a boy who sustained a BM relapse after 2 months of continued maintenance. There were two (7.7%) isolated BM relapses, one (3.8%) concomitant BM and overt testicular relapse, and two (7.7%) isolated overt testicular relapses. The latter occurred 7 and 14 months after completion of therapy. Assuming an occult testicular relapse rate of 10%, the expected number of testicular relapses in the nonbiopsied group was 2.6, and the observed number was 2. The rate of isolated overt testicular relapse was higher in the patients who did not undergo testicular biopsies than in boys with negative end-therapy biopsies (7.7% v 2.0%), but the difference was not statistically significant (P = .10) Comparisonof Outcomes The proportion of patients with negative testicular biopsies, positive biopsies, and without biop-
sies who experienced various end points after 3 years of CCR are presented in Table 7. The overall relapse rate, isolated BM relapse rate, and death rate were significantly higher in patients with occult TL than in patients with negative biopsies (P < .001) for each comparison. DFS after testicular biopsy was significantly better in patients without than with occult TL after 3 years of therapy (P < .001). The isolated BM relapse rate and death rates were significantly higher in patients with positive biopsies than in patients who did not undergo biopsies (P < .01). However, the difference in overall relapse rate (BM, CNS, testicular, extramedullary) in patients with positive biopsies and those without biopsies was not statistically significant (P > .10). There were no significant differences in the rates of various end points in the boys with negative biopsies and in those without biopsy data. Thus, the presence of occult TL after 3 years of CCR was associated with a greater risk of subsequent relapse and death despite the use of local radiation therapy and similar systemic reinduction and maintenance therapy. On the other hand, the small subset of boys who were not biopsied fared as well as those with negative biopsies. DISCUSSION
The clinical use of elective testicular biopsies before discontinuation of maintenance chemotherapy has been questioned in recent reports.' 9-21' 23 The low frequency of subsequent overt testicular relapse,"8 a high false-negative rate (low sensitivity),20 '23 and the efficacy of retrieval therapy for testicular relapse after treatment has been discontinued have been cited as arguments against the procedure. 23 The procedure would have merit if effective therapy in the boys with minimal residual TL would offer them
Table 7. Proportion of Patients Experiencing Various End Points Following Completion of 3-Year CCR
Group
Number
CCR
Relapse
Isolated BMR*
1. Negative biopsy 2. No biopsy 3. Positive biopsy
204 26 26
0.86* 0.811 0.62*1
0.12 0.191 0.39*1
0.06* 0.08: 0.34*
Isolated TR*
Isolated CNS
Concomitant BMR, TR, or CNSR
Death
0.02§ 0.08§ 0.00
0.02 0.00 0.04
0.02 0.04 0.00
0.05* 0.04t 0.23*
Abbreviations: BMR, bone marrow relapse; TR, testicular relapse; MR, marrow relapse, CNSR, central nervous system relapse. *P .1, group 2 v3. tP < .01, group 2 v3. §P> .10, group 1 v2.
Downloaded from ascopubs.org by RB DRAUGHTON LIBRARY on April 26, 2019 from 131.204.073.184 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
63
TESTICULAR BIOPSY IN CHILDHOOD ALL
the same DFS as males who do not have occult TL. The CCG-141 study substantiates that males with occult TL after 3 years of CCR had a significantly poorer DFS, hematologic remission duration, and overall survival than boys with negative testicular biopsies. Despite systemic reinduction, testicular radiation, and maintenance therapy as used initially in this trial, nine of 26 boys with occult TL relapsed in the BM and one relapsed in the CNS. Similar results were noted in the subsequent CCSG 160 series of protocols.2 5 Occult TL was identified in 72 of 718 (10%) after 2 years of maintenance therapy. Actuarial DFS 4 years after biopsy was 78% for boys with negative biopsies and 56% for boys with occult TL (P < .001). Fifteen of 646 boys developed overt TL after negative biopsies, a false-negative rate of 2.3%. These data support the findings in CCG-141 and indicate that occult TL is an adverse prognostic feature associated with an increased risk of BM relapse after therapy is discontinued. Overt testicular relapse during treatment occurs in 1% to 10% of boys',3,11,13,14 and fewer than 25% of these survive. In contrast, boys who develop overt TL after a negative end-therapy biopsy have an excellent outcome.7,12-15,20,21
Recently, 25 35 of 37 boys (94.6%) were reported to be in remission after re-treatment of offtherapy isolated testicular relapse. In an earlier CCSG trial in which end-therapy testicular biopsy was not required, Nesbit et a13 reported a survival of 56% in 16 boys (9.5% of those entered on the duration of therapy study) who developed isolated overt TL after 3 years of CCR. In comparison, the survival of boys with occult TL was 77% in CCG-141. The frequency of offtherapy isolated overt TL was 2.0%, significantly lower (P = .001) than in CCG-101/143. If only randomized patients in CCG-141 are evaluated, the frequency of isolated overt TL (two of 137) was 6.5 times less than in CCG-101/143 (16 of 168). Thus, performance of end-therapy testicular biopsies after 3 years of CCR was associated with a decreased frequency of isolated overt TL thereafter. Twenty-six boys on CCG-141 who did not have testicular biopsies were evaluable as a small control group. Only two developed isolated overt TL; both were successfully re-treated and remain
in CCR. These data and the results from the recent literature support the findings that current retrieval therapy is successful in 85% to 90% of boys who sustain isolated overt TL after 3 years of CCR. The frequency of occult TL in nearly 1,300 boys biopsied after 2 to 3 years of CCR has ranged from 0.9% to 40%.12-21 Isolated overt TL occurred in 0% to 10.7% (median, 5.7%) of the males who had negative biopsies.12-2' Thus, a negative testicular biopsy does not preclude a subsequent testicular relapse after the discontinuation of therapy. This purported relatively low sensitivity is in part related to the pitfalls associated with the histological determination of TL in submitted material. In CCG-141, the overall concordance of 92% between institutional and CPC review pathologists was encouraging, but full concordance in boys with positive testicular biopsies was 50% (13 of 26). In one other study of concordance after elective testicular biopsies BM relapse occurred in eight of 10 untreated boys in whom the institution biopsy reading was negative and the central review report was positive for occult TL.2 1 In CCG-141, outcome was similar in boys whose biopsies were read as positive either centrally or at the member institution. Stringent histologic criteria for occult TL are required. Nonstandardized surgical guidelines, bilateral versus unilateral biopsies, sampling error, inadequate or uninterpretable material, and difficulties inherent in distinguishing between testicular stromal cells, lymphocytes, and lymphoblasts were problems encountered in this and other studies.'1921 Positive testicular biopsies in most boys show a diffuse leukemic infiltration, but patchy focal involvement is observed as well. Histologically, occult TL shows interstitial, intertubular, or peritubular infiltrations similar to the 8 pattern recognized in most overt TL relapses.7' Rarely, involvement of the testicular tunic (tunica albuginea) or subtunic connective tissue occurs, making the diagnosis of occult leukemic infiltration particularly difficult. More definitive methods for distinguishing leukemic from nonleukemic cells are needed. Electron-microscopic evaluation may differentiate leukemic lymphoblasts from Leydig cells, primitive and peritubular fibroblasts, and normal lymphocytes in testicular infiltrates that are minimally cellular or show interstitial hyper-
Downloaded from ascopubs.org by RB DRAUGHTON LIBRARY on April 26, 2019 from 131.204.073.184 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
64
MILLER ET AL
cellularity. 26 However, terminal deoxynucleotidyl transferase (TdT) may have a limited value either as a marker for minimal residual leukemia or as a predictor of subsequent overt testicular relapse. 17 TdT-positive cells can be identified in histologically normal testicular biopsies. The absence of TdT-positive cells in histologically normal biopsies does not preclude a subsequent overt testicular relapse.' 8 The use of other markers (eg, monoclonal antibodies for the common acute lymphoblastic leukemia antigen 27) remains to be determined. Boys with occult TL had no distinguishing clinical prognostic features confirming the results of Wong et al.15 In contrast to the findings of Ortega et al,14 we found that an initial WBC count less than 20 x 109/1 rather than greater than 20 x 109/1 was associated with occult TL. Boys in the report by Nesbit et al28 who sustained isolated overt TL off-therapy had more splenomegaly at the time of original diagnosis, and to a lesser extent had less anemia (initial hemoglobin level > 7.5 gm/dL), and more thrombocytopenia. In our study none of the patients with occult TL had features of the leukemia/lymphoma syndrome as defined in CCG-141. Kim et a129 detected an association between T-cell acute lymphoblastic leukemia and early occult TL but there are no reported data correlating immunologic phenotype and late occult TL. Data on cell surface markers at diagnosis were unavailable in CCG-141. The clinical usefulness of elective testicular biopsies at completion of treatment would be demonstrated if boys with positive biopsy were found to have a higher frequency of recurrent disease after the discontinuation of treatment and thus might benefit from alternate therapy. 2 The results are conflicting. Tiedemann et al' reported no reduction in BM relapse rate after therapy was discontinued in boys with negative biopsies or without biopsies. In CCG-141, DFS was significantly better in randomized males
with negative biopsies enrolled in CCG-141 than in nonbiopsied males treated on CCG-101/ 143.2,3 Retreating patients with initial therapy did not improve the outlook in boys with occult TL, suggesting that more intensive treatment may be necessary to increase the DFS, eventfree, and overall survival of this clinically important subset. The flaws in these comparisons are obvious and a randomized trial with a concurrent nonbiopsied control group would be required to answer the question. On the basis of the testicular biopsy results, three groups of boys can be identified in CCG141. The first group (88%) were good responders to initial therapy, never had occult or overt TL, and had an excellent outcome; the second group (2%) were incomplete responders, developed overt TL after a negative biopsy and had an excellent outcome as well; the third group (10%) had an inadequate response to initial therapy, had occult TL after 3 years of CCR, and a poor outcome. Boys with occult TL appear to have more minimal-residual disease in other sites than boys without occult TL and have a worse prognosis because of their high risk of subsequent BM relapse, in part related to inadequate therapy. Indeed, occult TL is one of the few predictors of disease recurrence after 3 years of CCR. As such, end-therapy biopsies may be clinically useful, and may warrant further study. However, to identify the 10% or less at risk, all boys would require biopsy. In view of the decreasing incidence of overt TL off-therapy with more recent and intensive regimens,30-32 and the absence of clinical or laboratory features that identify boys at risk of having occult TL, such a strategy is questionable. It will be difficult to complete a prospective randomized trial designed to determine whether early diagnosis of occult TL and intensification of therapy would improve the ultimate outcome compared with retrieval therapy for boys who develop overt TL after therapy is discontinued.
APPENDIX
Principal Investigators, Childrens Cancer Study Group Institution Group Operations Office, University of Southern California Comprehensive Cancer Center, Los Angeles, CA
Investigator
Grant No.
Denman Hammond, MD John Weiner, DrPH Harland Sather, PhD Richard Sposto, PhD Mark Krailo, PhD Jonathan Buckley, MBBS, PhD Madeline Bauer, PhD
CA 13539
Downloaded from ascopubs.org by RB DRAUGHTON LIBRARY on April 26, 2019 from 131.204.073.184 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
65
TESTICULAR BIOPSY IN CHILDHOOD ALL APPENDIX
Principal Investigators, Childrens Cancer Study Group (Cont'd) Investigator
Institution University of Michigan Medical Center, Ann Arbor, MI University of California Medical Center, San Francisco, CA University of Wisconsin Hospital, Madison, WI Children's Hospital and Medical Center, Seattle, WA Rainbow Babies and Children's Hospital, Cleveland, OH Children's Hospital National Medical Center, Washington, DC Children's Memorial Hospital, Chicago, IL Children's Hospital of Los Angeles, Los Angeles, CA Children's Hospital of Columbus, Columbus, OH Babies' Hospital, New York, NY Children's Hospital of Pittsburgh, Pittsburgh, PA Doernbecher Memorial Hospital for Children, Portland, OR University of Minnesota Health Sciences Center, Minneapolis, MN University of Texas Health Sciences Center, San Antonio, TX Children's Hospital of Philadelphia, Philadelphia, Pennsylvania Memorial Sloan-Kettering Cancer Center, New York, NY James Whitcomb Riley Hospital for Children, Indianapolis, IN University of Utah Medical Center, Salt Lake City, UT Strong Memorial Hospital, Rochester, NY University of British Columbia, Vancouver, BC, Canada Harbor/UCLA & Miller's Children's Medical Center, Torrance and Long Beach, CA Children's Hospital of Denver, Denver, CO Izaak Walton Killam Hospital for Children Halifax, NS, Canada
Raymond Hutchinson, MD Arthur Ablin, MD Paul Gaynon, MD Ronald Chard, MD Susan Shurin, MD Gregory Reaman, MD Edward Baum, MD Jorge Ortega, MD Frederick Ruymann, MD Sergio Piomelli, MD Vincent Albo, MD Robert Neerhout, MD William Woods, MD Thomas Williams, MD Anna Meadows, MD Peter Steinherz, MD Robert Weetman, MD Richard O'Brien, MD Harvey Cohen, MD Paul Rogers, MD Jerry Finklestein, MD David Tubergen, MD Allan Pyesmany, MD
Grant No. CA 02971 CA 17829 CA 05436 CA 10382 CA 20320 CA 03888 CA 07431 CA 02649 CA 03750 CA 03526 CA 36015 CA 26044 CA 07306 CA 36004 CA 11796 CA 42764 CA 13809 CA 10198 CA 11174 CA 29013 CA 14560 CA 28851 -
REFERENCES 1. Miller DR, Leikin SL, Albo VC, et al: Prognostic factors and therapy in acute lymphoblastic leukemia of childhood: CCG 141. A report from Childrens Cancer Study Group. Cancer 51:1041-1049, 1983 2. Miller DR, Leikin SL, Albo VC, et al: Three versus five years of maintenance therapy are equivalent in childhood acute lymphoblastic leukemia: A report from the Childrens Cancer Study Group. J Clin Oncol 7:316-325, 1989 3. Nesbit ME, Robison LL, Ortega JA, et al: A randomized study of three years versus five years of chemotherapy in childhood acute lymphoblastic leukemia. J of Clin Oncol 1:308-316, 1983 4. George SL, Aur RJA, Mauer AM, et al: A reappraisal of the results of stopping therapy in childhood leukemia. New Eng J Med 300:269-273, 1979 5. Sather HN, Miller DR, Nesbit ME, et al: Differences in prognosis for boys and girls with acute lymphoblastic leukemia. Lancet 1:739-743, 1981 6. Land VJ, Berry DH, Herson J, et al: Long-term survival in childhood leukemia: "Late" relapses. Med Ped Oncol 7:19-24, 1979 7. Askin FB, Land VJ, Sullivan MP, et al: Occult testicular leukemia. Cancer 47:470-475, 1981 8. Eden OB, Hardisty RM, Innes EM, et al: Testicular disease in acute lymphoblastic leukemia of childhood. Br Med J 1:334-338, 1978 9. Finkelstein JZ, Dyment PG, Hammond GD: Leukemia infiltration of the testes during bone marrow remission. Pediatr 43:1042-1045, 1969 10. Kuo T-T, Tschang T-P, Chue J-Y: Testicular relapse in childhood acute leukemia during bone marrow remission. Cancer 38:2604-2612, 1976 11. Stoffel TJ, Nesbit ME, Levitt SH: Extramedullary involvement of the testes in childhood leukemia. Cancer 35:1203-1211, 1975
12. Tiedemann K, Chessels JM, Sandland RM: Isolated testicular relapse in boys with acute lymphoblastic leukemia: Treatment and outcome. Br Med J 285:1614-1616, 1982 13. Bowman WP, Aur RJA, Husto HO, et al: Isolated testicular relapse in acute lymphocytic leukemia of childhood: Categories and influence on survival. J Clin Oncol 2:924-929, 1984 14. Ortega JJ, Javier J, Toran N: Testicular infiltrates in children with acute lymphoblastic leukemia: A prospective study. Med Ped Oncol 12:386-393, 1984 15. Wong KY, Ballard ET, Strayer FH, et al: Clinical and occult testicular leukemia in long term survivors of acute lymphoblastic leukemia. J Pediatr 96:569-575, 1980 16. Mahoney DH Jr, Gonzales ET, Ferry JD, et al: Childhood acute leukemia: A search for occult extramedullary disease prior to discontinuation of therapy. Cancer 48:1964-1966, 1981 17. Chessells JM, Pincott J, Janossy J: TdT+ cells in routine testicular biopsy: Significance and relation to histology and clinical outcome. Br J Haematol 58:184-185, 1984 18. Chessells JM, Pincott JR, Daniels-Lake W: Terminal transferase positive cells in testicular biopsy specimens from boys with acute lymphoblastic leukemia. J Clin Pathol 39:1236-1240, 1986 19. Pui C-H, Dahl J, Bowman WP, et al: Elective testicular biopsy during chemotherapy for childhood leukemia is of no clinical value. Lancet 2:410-412, 1985 20. Hudson MM, Frankel LS, Mullins J, et al: Diagnostic value of surgical testicular biopsy after therapy for acute lymphocytic leukemia. J Pediatr 107:50-53, 1985 21. Nachman J, Coccia P, Bleyer A, et al: Overt and occult testicular leukemia (OTL) after two years of continuous remission: Diagnosis and treatment results. Blood 64: 170a, 1984 (abstr 591, suppl)
Downloaded from ascopubs.org by RB DRAUGHTON LIBRARY on April 26, 2019 from 131.204.073.184 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
66
MILLER ET AL
22. Sather HN, Coccia PF, Nesbit ME, et al: Disappearance of the predictive value of prognostic variables in childhood acute lymphoblastic leukemia. Cancer 48:370-376, 1981 23. Chessells JM: Acute leukemia in children. Clin Haematol 15:727-753, 1986 24. Peto R, Pike M, Armitage P, et al: Design and analysis of randomized clinical trials requiring prolonged observation of each patient: II. Analysis and examples. Br J Cancer 35:1-39, 1977 25. Nachman J, Coccia P, Bleyer W, et al: Treatment results for overt and occult testicular leukemia (OTL) diagnosed after 2 years of continuous remission. Blood 68:227a, 1986 (suppl) (abstract 788) 26. Prince FP, Palmer NF, Newton WA: Ultrastructural criteria for evaluating leukemic infiltration in prepubertal testicular biopsies. Pediatr Pathol 5:17-29, 1986 27. LeBien TW, McKenna RW, Abramson C, et al: Use of monoclonal antibodies, morphology and cytochemistry to
probe the cellular heterogeneity of acute leukemia and lymphoma. Cancer Research 41:4776-4780, 1981 28. Nesbit ME Jr, Robison LL, Ortega JA, et al: Testicular relapse in childhood acute lymphoblastic leukemia: Association with pre-treatment patient characteristics and treatment. Cancer 45:2009-2016, 1980 29. Kim TH, Hargreaves HK, Chan WC, et al: Sequential testicular biopsies in childhood acute lymphoblastic leukemia. Cancer 57:1028-1041, 1986 30. Clavell LA, Gelber RD, Cohen HJ, et al: Four agent induction and intensive asparaginase therapy for treatment of childhood acute lymphoblastic leukemia. N Eng J Med 315:657-661, 1986 31. Moe PJ, Seip M, Finne PH, et al: Methotrexate infusions in poor prognosis acute lymphoblastic leukemia in children. Med Pediatr Oncol 14:187-189, 1986 32. Gaynon P, Steinherz P, Bleyer WA, et al: Superiority of intensive therapy for children with acute lymphoblastic leukemia (ALL) and unfavorable presenting features. Lancet 2:921-924, 1988
Downloaded from ascopubs.org by RB DRAUGHTON LIBRARY on April 26, 2019 from 131.204.073.184 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
Lymphoblastic Leukemia: A Report From the Childrens Cancer Study Group By Denis R. Miller, Sanford L. Leikin, Vincent C. Albo, Nigel F. Palmer, Harland N. Sather, and G. Denman Hammond One of the objectives of Childrens Cancer Study Group (CCSG) study 141 (CCG-141) was to determine the frequency of occult testicular leukemia (TL) after 3
years of disease-free survival (DFS) and to retreat boys with occult TL to prolong their subsequent DFS. Of the 494 boys entered on study, 255 (51.6%) were in complete continuous remission (CCR) 3 years after entering remission and an additional eight were in CCR 3 years after localized extramedullary relapse and retreatment; 263 boys were eligible for testicular biopsy. Elective testicular biopsy was performed on 235 (89.4%) boys. Of the 204 (86.8%) boys with negative biopsies, 175 (85.8%) remained in CCR 10 to 12 years after diagnosis and 25 (12.3%) relapsed, 11 (44%)of whom died. Isolated overt TL occurred in four (2.0%) and all remained in CCR 22+ to 60+ months after re-treatment. Of the 26 boys with occult TL, 16 (62%) remained in CCR. Ten (38%) relapsed despite local testicular irradiation and systemic re-treatment;
C
URRENTLY, more than 50% of effectively treated children with acute lymphoblastic leukemia (ALL) are in continuous complete remission (CCR) 3 years or more after diagnosis. The value of "front-end" factors in determining the long-term prognosis of previously untreated children with ALL has attained universal acceptance and use. 1'2 However, until now it has not been possible to identify prognostic factors at the end of scheduled treatment that predict late relapse. Cessation of therapy after 3 years of CCR carries a risk of relapse of 10% to 20%, higher in males than in females."5 The reported rate of occult testicular leukemia (TL) after 2 to 3 years of CCR or overt TL relapse after therapy is discontinued is approximately 10%, and has been reported as high as 40% and as low as 0 . 9 %.6-21 It is difficult to predict which males are at risk of developing TL after cessation of treatment because, except for sex, other reported prognostic factors lose their ability to predict relapse in all sites after 2 years of CCR. 22 It is not certain whether occult TL is a sign of localized or widespread residual disease.12,19 The
six of the 10 died. Of the 26 boys who did not undergo biopsy, 21 (80.8%) remained in CCR; two (7.7%) developed isolated overt TL. DFS after testicular biopsy was significantly better in boys without occult TL (P = .001). Occult TL after 3 years of CCR represents aggressive minimal-residual disease and carries a worse prognosis than absence of TL. Initial treatment should be directed at obviating occult and overt testicular relapse. Conventional therapy as used in this study was suboptimal in preventing subsequent bone marrow (BM) relapse and death. If occult TR is identified during or at the end of planned therapy, a higher salvage rate may require intensified alternate therapy. As such, testicular biopsies may be clinically useful. Further investigation is limited by the relative rarity of, and the lack of identifying features in boys with occult TL. J Clin Oncol 8:57-66. @ 1990 by American Society of ClinicalOncology.
prognostic significance of isolated occult TL while on chemotherapy or at end-therapy has not been determined and continues to be the subject of controversy. 9-2 The clinical use of elective end-therapy testicular biopsies has been questioned by citing a high false-negative rate,2 3 a low incidence of occult TL,19, 20 and the success of treating isolated overt testicular relapse after therapy is discontinued.2 3 In a previously reported Childrens Cancer Study Group (CCSG) study (CCG-101/143), From the Lutheran General Children's Medical Center, Park Ridge, IL; Children's Hospital National Medical Center, Washington, DC; Children'sHospitalofPittsburgh, PA; Children'sHospitalof Columbus, OH; and the University of Southern CaliforniaSchool of Medicine, Los Angeles, CA. Submitted August 9, 1988; accepted August 25, 1989. Supported in part by the Division of Cancer Treatment, National Cancer Institute, National Institutes of Health, Department of Health and Human Services, Bethesda, MD. Address reprint requests to the Childrens Cancer Study Group, 199 North Lake Ave, Third Floor, Pasadena, CA 91101. © 1990 by American Society of Clinical Oncology. 0732-183X/90/0801-0008$3.00/0
Journalof Clinical Oncology, Vol 8, No 1 (January), 1990: pp 57-66
Downloaded from ascopubs.org by RB DRAUGHTON LIBRARY on April 26, 2019 from 131.204.073.184 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
57
58
MILLER ET AL
testicular biopsies were not performed before the cessation of therapy after either 3 years or 5 years of CCR. The overt testicular relapse rate off-therapy was 9.5%, 44% of whom died despite local and systemic retrieval therapy. In this study, CCG-141, open-wedge testicular biopsies were mandatory in all boys remaining in CCR for 3 years and were performed before the randomization to stop chemotherapy or continue chemotherapy for an additional 2 years. Objectives of the study were to determine the frequency of occult TL after 3 years of CCR, and to prolong subsequent recurrence-free survival in boys with occult TL by identifying and treating residual extramedullary disease. PATIENTS AND METHODS CCG-141 was opened in February 1975 and was closed to patient entry in February 1977. The details of the treatment regimens are published' and summarized in Fig 1. Briefly, patients less than 21 years of age were stratified according to low (< 20 x 109/1) or high (a 20 x 109/1) initial WBC count. Children with low initial WBC counts received regimen 1: induction with vincristine, prednisone, and asparaginase, CNS prophylaxis with 24 Gy cranial radiation and weekly intrathecal methotrexate for 6 weeks, and maintenance therapy for the first year with daily mercaptopurine, weekly methotrexate, and monthly pulses of vincristine and prednisone. Patients with high initial WBC counts were randomized to receive either regimen 1 or regimen 2, a more intensive program of induction, CNS prophylaxis, and maintenance Phoas I INDUCTION Day Io 42
"Low" REG I
therapy. After the first year of CCR all patients were maintained on daily mercaptopurine and weekly methotrexate. After 3 years of initial CCR or subsequent (secondary) CCR after localized extramedullary (CNS, testicular, skin, other) relapse, all males were to undergo bilateral openwedge biopsy. Biopsy material evaluated by each participating institution's pathologist was centrally reviewed by the CCSG Pathology Center (CPC) in Columbus, OH. Eligibility for the duration of therapy phase of CCG-141 required that patients have a negative biopsy as interpreted by the member institution and/or the CPC. A fully evaluable patient with a positive biopsy required protocol-prescribed local therapy and systemic reinduction and maintenance. Disposition and treatment were based upon the institution review in over 90% of the cases with positive biopsies. Boys with occult TL received local testicular radiation (20 to 24 Gy), reinduction therapy with vincristine, prednisone, and asparaginase, weekly intrathecal methotrexate for 6 weeks, and daily mercaptopurine, and weekly methotrexate for 3 years or until relapse. Patients with negative biopsies were eligible to be randomized to one of three regimens designed to determine the optimal duration of therapy. In regimen A, all chemotherapy was discontinued; in regimen B, patients received a 4-week course of reinduction therapy (vincristine, prednisone, and asparaginase) and then therapy was discontinued; in regimen C, maintenance therapy was continued for 2 more years and then stopped. Boys with equivocal biopsies were also randomized to regimens A, B, and C. Informed consent was obtained according to the investigator's institutional review board requirements. Comparisons between groups were made using the MantelPeto-Cox summary X2 (log-rank) method of life tableanalysis.24 The standard X2 statistic was used where appropri-
Phase II
Ph.ae III
Phase IV
Phot. V
CNS fx to 2 or onths
MAINTENANCE I to 12 months
MAINTENANCE II to 36 months from corpletiown of CNM thpy
MAINTENANCE III to 60 months from completion of CN th.Mry
PON
Phose VI OBSERVATION
VCR
"LOW GRONP" wBC 2o.0om And no Mediostinal wid*nin)
"Hig
"HIGH GROLPX
WaBC 2000 ond/o, Medilosinal
"w-dninq
M3 doa N · Polients _1
-m-,
OFF STUDY - 2001h
Fig 1.
Study schema, CCG-141.
Downloaded from ascopubs.org by RB DRAUGHTON LIBRARY on April 26, 2019 from 131.204.073.184 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
59
TESTICULAR BIOPSY IN CHILDHOOD ALL ate. The end points evaluated in tlhe life-table analysis included the following: (1) disease-fr ee survival (DFS) or time to first occurrence of bone marro' wrene of bone w, CNS, testicular, or testicular, row other extramedullary relapse; (2) occur rence of bone marrow (BM) relapse (> 25% blasts); (3) o,ccurrence of isolated symptomatic CNS relapse; (4) occurr ence of isolated overt testicular relapse; and (5) death.
tient permission for testicular biopsy was refused in 26 boys (9.9%) who then were not eligible for randomization. All boys who refused biopsy were
The data represent patient ffollow-up through June 1987. All patients are no,w 10 to 12 years postdiagnosis, and 3 to 9 years off-therapy. The median follow-up is 60+ mont ths from the time of testicular biopsy. Of the 494 boys entered on st:udy, 465 (94.1%) achieved a complete remission and 255 (51.6%) were in CCR 3 years after completing CNS therapy. Eight additional boys who had isolated extramedullary relapses during primary maintenance therapy (CNS, five; tes ticular, two; gingiva/palate, one) received prrotocol-prescribed therapy for an additional 36 ± 3 months and were eligible to undergo biopsy (Table 1). Open-wedge testicular bio psies were performed in 235 (89.4%) of the 2( 63 eligible boys, of whom 196 (83.4%) had bilater;al biopsies and 14 (6.0%) had unilateral biopsies;. Information regarding unilaterality or bilate :rality of biopsies was unavailable in 25 (10.6%) ).Parental or pa-
followed on the duration of therapy study. Of all the males biopsied, 204 (86.8%) had no testicular involvement and 31 (13.2%) had occult TL. Of the 230 males enrolled in the duration of therapy study, 137 (59.6%) were randomly assigned to one of the three study regimens; of the 93 boys (40.4%) who were assigned nonrandomly, 67 had negative biopsies, 24 were not biopsied, and two had discordant positive biopsies. The nonrandom assignment rate of 40% is in accord with other CCSG clinical trials and reflects primarily parental reference, and also was in accord with informed consent. The DFS rates in boys with negative testicular biopsies who were assigned randomly (86.7%, observed: expected ratio, 0.83) and nonrandomly (85.6% observed: expected rate, 0.88) were virtually identical. Thus, all boys with negative biopsies were analyzed together. Thirty-one boys had isolated occult TL (13.2% of those biopsied). Five boys with occult TL were excluded from further analysis, three with discordant results and two with biopsies performed before or after the protocol-prescribed time point. The protocol stipulated that biopsies be per-
Table 1. Testicular Biopsy Study: Summary Data
formed within 3 to 5 months of the predeter-
RESULTS
Patients
rNumber
Entered on study CCG-141 Completed induction with CR In CCR at 36 -- 6 months after CNS therapy In secondary CCR 36 ± 6 months after isolated extramedullary relapse* Eligible for testicular biopsy No biopsy performed Biopsied Negative biopsy Positive biopsy Protocol compliant Protocol deviations Entered on duration of therapy study Randomized Nonrandomized
494
Percent\
100
465
94.1
255
51.6
8*
1.6*
263
53.2
26 235 204 31
9.9 89.4
26
11.1
86.8
13.2
230
mined completion date of primary (3 months) or A
+
LL""'" Y 1111111111 ocs naJ---nh)maneaneter
lllnllllllnlll.l
Tt.h
Illnllv
II
biopsy was performed too early in one case (29
months) and too late (44 months) in another. Three patients with discordant biopsy (institu-
tion negative-CPC positive, 2; institution positiveCPC negative, 1) were entered on the duration of therapy study and had therapy discontinued, one randomly and two nonrandomly. Thus, there were 26 boys with isolated occult TL treated per protocol and fully evaluated for fate after biopsy. Two boys with occult TL were treated with orchidectomy of the involved testis, systemic reinduction, and maintenance therapy and were radiation was included in the analysis. Testicular
137
59.6
not given.
93
40.4
The clinical characteristics of the 31 bovs with isolated occult TL, 204 boys with negative biopsies, and the 26 who were not biopsied are j
Abbreviation: CR, complete response. *Isolated extramedullary relapses: CNIS,five; overt testicular, two; gum/palate ulcer, one; (two of the eigght had occult ThL). biopsy (29 months), one; late biopsy (44 months), one; discordant review, entered on duration of therapy study, three
"tEorly
(therapy discontinued).
.......
compared in Table 2. Only one boy with occult TL had an initial WBC count greater than 50 x
109/1 and none had features of the leukemia/
Downloaded from ascopubs.org by RB DRAUGHTON LIBRARY on April 26, 2019 from 131.204.073.184 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
60
MILLER ET AL Table 2. Clinical Characteristics of 235 Boys Undergoing Testicular Biopsy After 3 Years CCR and 26 Boys Not Biopsied
Characteristics Initial WBC count - 20 x 10'/L > 20 x 109/L Age < 12 months 13 to 120 months > 120 months Prognostic group Good Average Poor FAB classification L1 L2 Unknown IG level Normal Decreased Unknown Day-14 BM M1 M2/M3 Unknown Hemoglobin < 10 gm/dL - 10 gm/dL Splenomegaly, marked/moderate Mediastinal mass, present Leukemia/lymphoma syndrome
Occult Testicular
Negative Biopsy
Leukemia (N = 31)*
(N = 204)
No Biopsy (N = 26)
N
%
N
%
N
5
26 5
83.9 16.1
153 51
75.0 25.0
20 6
76.9 23.1
0 28 3
0.0 90.3 9.7
4 166 34
2.0 81.4 16.6
0 22 4
0.0 84.6 15.4
11 19 1
35.5 61.3 3.2
55 125 24
27.0 61.3 11.7
10 12 4
38.5 46.2 15.4
27 2 2
87.0 6.5 6.5
163 24 17
79.9 11.8 8.3
21 3 2
80.8 11.5 7.7
21 4 6
67.7 12.9 19.4
129 45 30
63.2 22.1 14.7
13 3 10
50.0 11.5 38.5
18 6 7
58.1 19.3 22.6
120 53 31
58.8 26.0 15.2
17 4 5
65.4 15.4 19.2
28 3 13 0 0
90.3 9.7 41.9 0.0 0.0
167 37 87 11 15
81.9 18.1 42.6 5.3 7.3
22 4 9 1 1
84.6 15.4 34.6 3.8 3.8
Abbreviations: CCR, complete continuous remission; FAB, French-American-British; IG, immunoglobin. *Includes five ineligible boys, three with discordant positive biopsies entered on duration of therapy study, one with early biopsy (29 months), one with late biopsy (44 months). "fPrognostic group definitions: good, initial WBC count < 10,000//uL, age 3 to 7; poor, initial WBC count n 50,000/tL; average, all other patients. tSee text for definition.
lymphoma syndrome as defined in this study1 (presence of three of the five following features: splenomegaly below the umbilicus, marked lymphodenopathy [visible, 3 cm single node or 5 cm mass], mediastinal mass, hemoglobin _ 10 gm/ dl, or WBC count > 50 x 109/1). The distribution by initial WBC count, age, prognostic group, lymphoblast morphology, immunoglobulin levels, results of the day-14 BM, hemoglobin level, degree of splenomegaly and lymphadenopathy, presence of mediastinal mass, or presence of leukemia/lymphoma syndrome was not significantly different in boys with occult TL compared with those with negative biopsies or those not biopsied. Cell surface marker and chromosomal analyses were not required in CCG-141. Correla-
tions between immunologic phenotype, karyotype, and occult TL were not determined. ConcordanceStudy Of the 235 testicular biopsies performed, 172 (72.6%) were reviewed by the CPC, of which 163 (94.7%) were assessable (Table 3). The review was concordant with the member institution in 150 biopsies (92.0%) and discordant in 13 (8.0%). Unilateral rather than bilateral involvement was noted in only two of 144 (1.4%) bilateral biopsies reviewed by the CPC. Of the 31 boys with isolated occult TL, five biopsy samples that were positive at the member institution were not reviewed centrally. Of the remaining 26 biopsies, 13 biopsies were fully concordant, four were
Downloaded from ascopubs.org by RB DRAUGHTON LIBRARY on April 26, 2019 from 131.204.073.184 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
61
TESTICULAR BIOPSY IN CHILDHOOD ALL Table 3. Testicular Biopsy: Concordance Study Number Males biopsied
235
Percent 100
Not reviewed/inevaluable by CPC Institution +t Institution -t Assessable by CPC Concordant review Institution +/CPC Institution -/CPC Discordant review Institution +/CPC Institution -/CPC
+ +
74* 7 67 163 150 13 137 13 4 9
31.2 2.9 28.3 68.8 92.0t (63.8)§ (5.5) (58.3) 8.0f (5.5) (1.7) (3.8)
*Includes nine specimens reviewed by CPC, but inassessable. t +, positive; -, negative. *Percent of assessable specimens. §Number in parentheses equals percent of total biopsy specimens.
interpreted as positive at the member institution and negative at the CPC, and nine were considered negative at the member institution and positive by the CPC. Fate After TesticularBiopsy Of the 204 boys with negative testicular biopsies (Table 4), 175 (85.8%) remained in CCR, 25 (12.3%) relapsed, and four (2.0%) died of nonleukemic causes (three with malignant brain tumors, one of disseminated varicella). Isolated BM relapse occurred in 13 boys, 1 to 36 months (median, 11 months), following testicular biopsy and nine have died. Concomitant BM and overt TL occurred in four boys, nine to 47 months after testicular biopsy, and one has died. Isolated overt TL occurred in four boys with prior negative biopsies, a "false-negative" rate of 2.0%. The Table 4. Fate After Negative Biopsy Number Negative CCR Relapse BM BM/overt testicular Isolated overt testicular Isolated CNS Alive after relapse Dead after relapse Death unrelated to leukemia
204 175 25 13 4 4 4t 14 11 4*t
Percent 200 85.8 12.3 6.4 2.0 2.0 2.0 56.0 44.0 2.0
*Brain tumors, three; disseminated varicella, one. tlncludes one patient with CNS relapse 16 months after testicular biopsy, death due to high-grade astrocytoma 25 months after biopsy.
overt testicular relapses occurred from 14 to 49 months after testicular biopsy, but all four patients remain in CCR 22+ to 60+ months after protocol-prescribed treatment. Isolated CNS relapses occurred in four boys, 3 to 56 months after biopsy. Three remain in CCR after retreatment. One, who developed a right fronto-parietal lesion diagnosed by needle biopsy as "CNS leukemia" 16 months after testicular biopsy, died of a high-grade astrocytoma 9 months later. The total relapse rate was significantly higher in boys with negative testicular biopsies who had chemotherapy randomly or nonrandomly discontinued after 3 years (regimen A, 13 of 84, 15.5%) compared with boys who were treated for 5 years (regimen C, four of 64, 5.8%, P < .05). The difference was not significant when only randomly-assigned patients were compared. The fate of boys with isolated occult TL is shown in Table 5. Of the 26 boys with occult TL, 16 (61.5%) remained in CCR after local radiation (or orchidectomy in two), systemic reinduction, CNS prophylaxis with intrathecal methotrexate, and 2 to 3 years of maintenance therapy. The median duration of follow-up after testicular biopsy was 60+ months (range, 17 to 73+ months). Ten (38.5%) patients with occult TL relapsed: nine sustained isolated BM relapses 12 to 30 months (median, 17 months) after occult TL, six of whom have died. One patient developed an isolated CNS relapse 23 months after detection of occult TL and remains in complete remission 41 + months. At the time of the latest update all boys who had occult TL and were re-treated were off therapy. The relapse rate did not differ in patients in whom the determination of a positive biopsy was made by the member institution or by the central review pathologist; relapses were evenly distributed between institution-positive and CPC-positive biopsies. Of the 26 patients who did not have testicular Table 5. Fate After Positive Testicular Biopsy and Re-treatment
Positive biopsy CCR Relapse BM CNS Dead after relapse
Number
Percent
26 16 10 9 1 6
100 61.5 38.5 34.6 3.9 60
Downloaded from ascopubs.org by RB DRAUGHTON LIBRARY on April 26, 2019 from 131.204.073.184 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
62
MILLER ET AL
Table 6. Fate After No Biopsy (Refusal/Protocol Violation) Number
Percent
26 21 5 2 2 0 1 1 4
100 80.8 19.2 7.7 7.7
No biopsy CCR Relapse Isolated BM Isolated testicular Isolated CNS BM plus testicular Dead after relapse Alive after relapse
3.8 20.0 80.0
biopsies 21 (80.8%) remained in CCR 22+ to 84+ months (median, 57 months) from the completion date of 3 years of maintenance therapy (Table 6). Five patients (19.2%) relapsed, in three of whom therapy was nonrandomly discontinued and in two of whom maintenance therapy was continued for an additional 2 years. Four of the five relapses occurred after therapy had been discontinued. The only death occurred in a boy who sustained a BM relapse after 2 months of continued maintenance. There were two (7.7%) isolated BM relapses, one (3.8%) concomitant BM and overt testicular relapse, and two (7.7%) isolated overt testicular relapses. The latter occurred 7 and 14 months after completion of therapy. Assuming an occult testicular relapse rate of 10%, the expected number of testicular relapses in the nonbiopsied group was 2.6, and the observed number was 2. The rate of isolated overt testicular relapse was higher in the patients who did not undergo testicular biopsies than in boys with negative end-therapy biopsies (7.7% v 2.0%), but the difference was not statistically significant (P = .10) Comparisonof Outcomes The proportion of patients with negative testicular biopsies, positive biopsies, and without biop-
sies who experienced various end points after 3 years of CCR are presented in Table 7. The overall relapse rate, isolated BM relapse rate, and death rate were significantly higher in patients with occult TL than in patients with negative biopsies (P < .001) for each comparison. DFS after testicular biopsy was significantly better in patients without than with occult TL after 3 years of therapy (P < .001). The isolated BM relapse rate and death rates were significantly higher in patients with positive biopsies than in patients who did not undergo biopsies (P < .01). However, the difference in overall relapse rate (BM, CNS, testicular, extramedullary) in patients with positive biopsies and those without biopsies was not statistically significant (P > .10). There were no significant differences in the rates of various end points in the boys with negative biopsies and in those without biopsy data. Thus, the presence of occult TL after 3 years of CCR was associated with a greater risk of subsequent relapse and death despite the use of local radiation therapy and similar systemic reinduction and maintenance therapy. On the other hand, the small subset of boys who were not biopsied fared as well as those with negative biopsies. DISCUSSION
The clinical use of elective testicular biopsies before discontinuation of maintenance chemotherapy has been questioned in recent reports.' 9-21' 23 The low frequency of subsequent overt testicular relapse,"8 a high false-negative rate (low sensitivity),20 '23 and the efficacy of retrieval therapy for testicular relapse after treatment has been discontinued have been cited as arguments against the procedure. 23 The procedure would have merit if effective therapy in the boys with minimal residual TL would offer them
Table 7. Proportion of Patients Experiencing Various End Points Following Completion of 3-Year CCR
Group
Number
CCR
Relapse
Isolated BMR*
1. Negative biopsy 2. No biopsy 3. Positive biopsy
204 26 26
0.86* 0.811 0.62*1
0.12 0.191 0.39*1
0.06* 0.08: 0.34*
Isolated TR*
Isolated CNS
Concomitant BMR, TR, or CNSR
Death
0.02§ 0.08§ 0.00
0.02 0.00 0.04
0.02 0.04 0.00
0.05* 0.04t 0.23*
Abbreviations: BMR, bone marrow relapse; TR, testicular relapse; MR, marrow relapse, CNSR, central nervous system relapse. *P .1, group 2 v3. tP < .01, group 2 v3. §P> .10, group 1 v2.
Downloaded from ascopubs.org by RB DRAUGHTON LIBRARY on April 26, 2019 from 131.204.073.184 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
63
TESTICULAR BIOPSY IN CHILDHOOD ALL
the same DFS as males who do not have occult TL. The CCG-141 study substantiates that males with occult TL after 3 years of CCR had a significantly poorer DFS, hematologic remission duration, and overall survival than boys with negative testicular biopsies. Despite systemic reinduction, testicular radiation, and maintenance therapy as used initially in this trial, nine of 26 boys with occult TL relapsed in the BM and one relapsed in the CNS. Similar results were noted in the subsequent CCSG 160 series of protocols.2 5 Occult TL was identified in 72 of 718 (10%) after 2 years of maintenance therapy. Actuarial DFS 4 years after biopsy was 78% for boys with negative biopsies and 56% for boys with occult TL (P < .001). Fifteen of 646 boys developed overt TL after negative biopsies, a false-negative rate of 2.3%. These data support the findings in CCG-141 and indicate that occult TL is an adverse prognostic feature associated with an increased risk of BM relapse after therapy is discontinued. Overt testicular relapse during treatment occurs in 1% to 10% of boys',3,11,13,14 and fewer than 25% of these survive. In contrast, boys who develop overt TL after a negative end-therapy biopsy have an excellent outcome.7,12-15,20,21
Recently, 25 35 of 37 boys (94.6%) were reported to be in remission after re-treatment of offtherapy isolated testicular relapse. In an earlier CCSG trial in which end-therapy testicular biopsy was not required, Nesbit et a13 reported a survival of 56% in 16 boys (9.5% of those entered on the duration of therapy study) who developed isolated overt TL after 3 years of CCR. In comparison, the survival of boys with occult TL was 77% in CCG-141. The frequency of offtherapy isolated overt TL was 2.0%, significantly lower (P = .001) than in CCG-101/143. If only randomized patients in CCG-141 are evaluated, the frequency of isolated overt TL (two of 137) was 6.5 times less than in CCG-101/143 (16 of 168). Thus, performance of end-therapy testicular biopsies after 3 years of CCR was associated with a decreased frequency of isolated overt TL thereafter. Twenty-six boys on CCG-141 who did not have testicular biopsies were evaluable as a small control group. Only two developed isolated overt TL; both were successfully re-treated and remain
in CCR. These data and the results from the recent literature support the findings that current retrieval therapy is successful in 85% to 90% of boys who sustain isolated overt TL after 3 years of CCR. The frequency of occult TL in nearly 1,300 boys biopsied after 2 to 3 years of CCR has ranged from 0.9% to 40%.12-21 Isolated overt TL occurred in 0% to 10.7% (median, 5.7%) of the males who had negative biopsies.12-2' Thus, a negative testicular biopsy does not preclude a subsequent testicular relapse after the discontinuation of therapy. This purported relatively low sensitivity is in part related to the pitfalls associated with the histological determination of TL in submitted material. In CCG-141, the overall concordance of 92% between institutional and CPC review pathologists was encouraging, but full concordance in boys with positive testicular biopsies was 50% (13 of 26). In one other study of concordance after elective testicular biopsies BM relapse occurred in eight of 10 untreated boys in whom the institution biopsy reading was negative and the central review report was positive for occult TL.2 1 In CCG-141, outcome was similar in boys whose biopsies were read as positive either centrally or at the member institution. Stringent histologic criteria for occult TL are required. Nonstandardized surgical guidelines, bilateral versus unilateral biopsies, sampling error, inadequate or uninterpretable material, and difficulties inherent in distinguishing between testicular stromal cells, lymphocytes, and lymphoblasts were problems encountered in this and other studies.'1921 Positive testicular biopsies in most boys show a diffuse leukemic infiltration, but patchy focal involvement is observed as well. Histologically, occult TL shows interstitial, intertubular, or peritubular infiltrations similar to the 8 pattern recognized in most overt TL relapses.7' Rarely, involvement of the testicular tunic (tunica albuginea) or subtunic connective tissue occurs, making the diagnosis of occult leukemic infiltration particularly difficult. More definitive methods for distinguishing leukemic from nonleukemic cells are needed. Electron-microscopic evaluation may differentiate leukemic lymphoblasts from Leydig cells, primitive and peritubular fibroblasts, and normal lymphocytes in testicular infiltrates that are minimally cellular or show interstitial hyper-
Downloaded from ascopubs.org by RB DRAUGHTON LIBRARY on April 26, 2019 from 131.204.073.184 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
64
MILLER ET AL
cellularity. 26 However, terminal deoxynucleotidyl transferase (TdT) may have a limited value either as a marker for minimal residual leukemia or as a predictor of subsequent overt testicular relapse. 17 TdT-positive cells can be identified in histologically normal testicular biopsies. The absence of TdT-positive cells in histologically normal biopsies does not preclude a subsequent overt testicular relapse.' 8 The use of other markers (eg, monoclonal antibodies for the common acute lymphoblastic leukemia antigen 27) remains to be determined. Boys with occult TL had no distinguishing clinical prognostic features confirming the results of Wong et al.15 In contrast to the findings of Ortega et al,14 we found that an initial WBC count less than 20 x 109/1 rather than greater than 20 x 109/1 was associated with occult TL. Boys in the report by Nesbit et al28 who sustained isolated overt TL off-therapy had more splenomegaly at the time of original diagnosis, and to a lesser extent had less anemia (initial hemoglobin level > 7.5 gm/dL), and more thrombocytopenia. In our study none of the patients with occult TL had features of the leukemia/lymphoma syndrome as defined in CCG-141. Kim et a129 detected an association between T-cell acute lymphoblastic leukemia and early occult TL but there are no reported data correlating immunologic phenotype and late occult TL. Data on cell surface markers at diagnosis were unavailable in CCG-141. The clinical usefulness of elective testicular biopsies at completion of treatment would be demonstrated if boys with positive biopsy were found to have a higher frequency of recurrent disease after the discontinuation of treatment and thus might benefit from alternate therapy. 2 The results are conflicting. Tiedemann et al' reported no reduction in BM relapse rate after therapy was discontinued in boys with negative biopsies or without biopsies. In CCG-141, DFS was significantly better in randomized males
with negative biopsies enrolled in CCG-141 than in nonbiopsied males treated on CCG-101/ 143.2,3 Retreating patients with initial therapy did not improve the outlook in boys with occult TL, suggesting that more intensive treatment may be necessary to increase the DFS, eventfree, and overall survival of this clinically important subset. The flaws in these comparisons are obvious and a randomized trial with a concurrent nonbiopsied control group would be required to answer the question. On the basis of the testicular biopsy results, three groups of boys can be identified in CCG141. The first group (88%) were good responders to initial therapy, never had occult or overt TL, and had an excellent outcome; the second group (2%) were incomplete responders, developed overt TL after a negative biopsy and had an excellent outcome as well; the third group (10%) had an inadequate response to initial therapy, had occult TL after 3 years of CCR, and a poor outcome. Boys with occult TL appear to have more minimal-residual disease in other sites than boys without occult TL and have a worse prognosis because of their high risk of subsequent BM relapse, in part related to inadequate therapy. Indeed, occult TL is one of the few predictors of disease recurrence after 3 years of CCR. As such, end-therapy biopsies may be clinically useful, and may warrant further study. However, to identify the 10% or less at risk, all boys would require biopsy. In view of the decreasing incidence of overt TL off-therapy with more recent and intensive regimens,30-32 and the absence of clinical or laboratory features that identify boys at risk of having occult TL, such a strategy is questionable. It will be difficult to complete a prospective randomized trial designed to determine whether early diagnosis of occult TL and intensification of therapy would improve the ultimate outcome compared with retrieval therapy for boys who develop overt TL after therapy is discontinued.
APPENDIX
Principal Investigators, Childrens Cancer Study Group Institution Group Operations Office, University of Southern California Comprehensive Cancer Center, Los Angeles, CA
Investigator
Grant No.
Denman Hammond, MD John Weiner, DrPH Harland Sather, PhD Richard Sposto, PhD Mark Krailo, PhD Jonathan Buckley, MBBS, PhD Madeline Bauer, PhD
CA 13539
Downloaded from ascopubs.org by RB DRAUGHTON LIBRARY on April 26, 2019 from 131.204.073.184 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
65
TESTICULAR BIOPSY IN CHILDHOOD ALL APPENDIX
Principal Investigators, Childrens Cancer Study Group (Cont'd) Investigator
Institution University of Michigan Medical Center, Ann Arbor, MI University of California Medical Center, San Francisco, CA University of Wisconsin Hospital, Madison, WI Children's Hospital and Medical Center, Seattle, WA Rainbow Babies and Children's Hospital, Cleveland, OH Children's Hospital National Medical Center, Washington, DC Children's Memorial Hospital, Chicago, IL Children's Hospital of Los Angeles, Los Angeles, CA Children's Hospital of Columbus, Columbus, OH Babies' Hospital, New York, NY Children's Hospital of Pittsburgh, Pittsburgh, PA Doernbecher Memorial Hospital for Children, Portland, OR University of Minnesota Health Sciences Center, Minneapolis, MN University of Texas Health Sciences Center, San Antonio, TX Children's Hospital of Philadelphia, Philadelphia, Pennsylvania Memorial Sloan-Kettering Cancer Center, New York, NY James Whitcomb Riley Hospital for Children, Indianapolis, IN University of Utah Medical Center, Salt Lake City, UT Strong Memorial Hospital, Rochester, NY University of British Columbia, Vancouver, BC, Canada Harbor/UCLA & Miller's Children's Medical Center, Torrance and Long Beach, CA Children's Hospital of Denver, Denver, CO Izaak Walton Killam Hospital for Children Halifax, NS, Canada
Raymond Hutchinson, MD Arthur Ablin, MD Paul Gaynon, MD Ronald Chard, MD Susan Shurin, MD Gregory Reaman, MD Edward Baum, MD Jorge Ortega, MD Frederick Ruymann, MD Sergio Piomelli, MD Vincent Albo, MD Robert Neerhout, MD William Woods, MD Thomas Williams, MD Anna Meadows, MD Peter Steinherz, MD Robert Weetman, MD Richard O'Brien, MD Harvey Cohen, MD Paul Rogers, MD Jerry Finklestein, MD David Tubergen, MD Allan Pyesmany, MD
Grant No. CA 02971 CA 17829 CA 05436 CA 10382 CA 20320 CA 03888 CA 07431 CA 02649 CA 03750 CA 03526 CA 36015 CA 26044 CA 07306 CA 36004 CA 11796 CA 42764 CA 13809 CA 10198 CA 11174 CA 29013 CA 14560 CA 28851 -
REFERENCES 1. Miller DR, Leikin SL, Albo VC, et al: Prognostic factors and therapy in acute lymphoblastic leukemia of childhood: CCG 141. A report from Childrens Cancer Study Group. Cancer 51:1041-1049, 1983 2. Miller DR, Leikin SL, Albo VC, et al: Three versus five years of maintenance therapy are equivalent in childhood acute lymphoblastic leukemia: A report from the Childrens Cancer Study Group. J Clin Oncol 7:316-325, 1989 3. Nesbit ME, Robison LL, Ortega JA, et al: A randomized study of three years versus five years of chemotherapy in childhood acute lymphoblastic leukemia. J of Clin Oncol 1:308-316, 1983 4. George SL, Aur RJA, Mauer AM, et al: A reappraisal of the results of stopping therapy in childhood leukemia. New Eng J Med 300:269-273, 1979 5. Sather HN, Miller DR, Nesbit ME, et al: Differences in prognosis for boys and girls with acute lymphoblastic leukemia. Lancet 1:739-743, 1981 6. Land VJ, Berry DH, Herson J, et al: Long-term survival in childhood leukemia: "Late" relapses. Med Ped Oncol 7:19-24, 1979 7. Askin FB, Land VJ, Sullivan MP, et al: Occult testicular leukemia. Cancer 47:470-475, 1981 8. Eden OB, Hardisty RM, Innes EM, et al: Testicular disease in acute lymphoblastic leukemia of childhood. Br Med J 1:334-338, 1978 9. Finkelstein JZ, Dyment PG, Hammond GD: Leukemia infiltration of the testes during bone marrow remission. Pediatr 43:1042-1045, 1969 10. Kuo T-T, Tschang T-P, Chue J-Y: Testicular relapse in childhood acute leukemia during bone marrow remission. Cancer 38:2604-2612, 1976 11. Stoffel TJ, Nesbit ME, Levitt SH: Extramedullary involvement of the testes in childhood leukemia. Cancer 35:1203-1211, 1975
12. Tiedemann K, Chessels JM, Sandland RM: Isolated testicular relapse in boys with acute lymphoblastic leukemia: Treatment and outcome. Br Med J 285:1614-1616, 1982 13. Bowman WP, Aur RJA, Husto HO, et al: Isolated testicular relapse in acute lymphocytic leukemia of childhood: Categories and influence on survival. J Clin Oncol 2:924-929, 1984 14. Ortega JJ, Javier J, Toran N: Testicular infiltrates in children with acute lymphoblastic leukemia: A prospective study. Med Ped Oncol 12:386-393, 1984 15. Wong KY, Ballard ET, Strayer FH, et al: Clinical and occult testicular leukemia in long term survivors of acute lymphoblastic leukemia. J Pediatr 96:569-575, 1980 16. Mahoney DH Jr, Gonzales ET, Ferry JD, et al: Childhood acute leukemia: A search for occult extramedullary disease prior to discontinuation of therapy. Cancer 48:1964-1966, 1981 17. Chessells JM, Pincott J, Janossy J: TdT+ cells in routine testicular biopsy: Significance and relation to histology and clinical outcome. Br J Haematol 58:184-185, 1984 18. Chessells JM, Pincott JR, Daniels-Lake W: Terminal transferase positive cells in testicular biopsy specimens from boys with acute lymphoblastic leukemia. J Clin Pathol 39:1236-1240, 1986 19. Pui C-H, Dahl J, Bowman WP, et al: Elective testicular biopsy during chemotherapy for childhood leukemia is of no clinical value. Lancet 2:410-412, 1985 20. Hudson MM, Frankel LS, Mullins J, et al: Diagnostic value of surgical testicular biopsy after therapy for acute lymphocytic leukemia. J Pediatr 107:50-53, 1985 21. Nachman J, Coccia P, Bleyer A, et al: Overt and occult testicular leukemia (OTL) after two years of continuous remission: Diagnosis and treatment results. Blood 64: 170a, 1984 (abstr 591, suppl)
Downloaded from ascopubs.org by RB DRAUGHTON LIBRARY on April 26, 2019 from 131.204.073.184 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
66
MILLER ET AL
22. Sather HN, Coccia PF, Nesbit ME, et al: Disappearance of the predictive value of prognostic variables in childhood acute lymphoblastic leukemia. Cancer 48:370-376, 1981 23. Chessells JM: Acute leukemia in children. Clin Haematol 15:727-753, 1986 24. Peto R, Pike M, Armitage P, et al: Design and analysis of randomized clinical trials requiring prolonged observation of each patient: II. Analysis and examples. Br J Cancer 35:1-39, 1977 25. Nachman J, Coccia P, Bleyer W, et al: Treatment results for overt and occult testicular leukemia (OTL) diagnosed after 2 years of continuous remission. Blood 68:227a, 1986 (suppl) (abstract 788) 26. Prince FP, Palmer NF, Newton WA: Ultrastructural criteria for evaluating leukemic infiltration in prepubertal testicular biopsies. Pediatr Pathol 5:17-29, 1986 27. LeBien TW, McKenna RW, Abramson C, et al: Use of monoclonal antibodies, morphology and cytochemistry to
probe the cellular heterogeneity of acute leukemia and lymphoma. Cancer Research 41:4776-4780, 1981 28. Nesbit ME Jr, Robison LL, Ortega JA, et al: Testicular relapse in childhood acute lymphoblastic leukemia: Association with pre-treatment patient characteristics and treatment. Cancer 45:2009-2016, 1980 29. Kim TH, Hargreaves HK, Chan WC, et al: Sequential testicular biopsies in childhood acute lymphoblastic leukemia. Cancer 57:1028-1041, 1986 30. Clavell LA, Gelber RD, Cohen HJ, et al: Four agent induction and intensive asparaginase therapy for treatment of childhood acute lymphoblastic leukemia. N Eng J Med 315:657-661, 1986 31. Moe PJ, Seip M, Finne PH, et al: Methotrexate infusions in poor prognosis acute lymphoblastic leukemia in children. Med Pediatr Oncol 14:187-189, 1986 32. Gaynon P, Steinherz P, Bleyer WA, et al: Superiority of intensive therapy for children with acute lymphoblastic leukemia (ALL) and unfavorable presenting features. Lancet 2:921-924, 1988
Downloaded from ascopubs.org by RB DRAUGHTON LIBRARY on April 26, 2019 from 131.204.073.184 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
