Radiation Oncology/SCLC/PCI

Prophylactic Cranial Irradiation (PCI)

Natural History

• Risk of brain mets at 2 years is clinically ~50%, and ~80% in autopsy series.

• Wisconsin, 1981 (1974-1979) PMID 6268295 -- "Risk of brain metastasis from small cell carcinoma of the lung related to length of survival and prophylactic irradiation." (Komaki R, Cancer Treat Rep. 1981 Sep-Oct;65(9-10):811-4.)
  • Retrospective. 131 patients, 57 patients +PCI, 74 patients -PCI. Different periods of observation
  • Calculated clinical failure: 12 months: +PCI 11% vs. -PCI 28%; 24 months +PCI 11% vs. -PCI 58% (SS)

• 1979 PMID 227582 -- "CNS metastases in small cell bronchogenic carcinoma: increasing frequency and changing pattern with lengthening survival." (Nugent JL, Cancer. 1979 Nov;44(5):1885-93.)
  • Autopsy series: 2-year CNS mets in 80% patients

Treatment Overview

• PCI was first proposed in 1973 due to high rate of developing brain mets, and their unsatisfactory control with whole brain RT
• Two meta-analyses showed benefit in terms of brain mets incidence (50% reduction) and survival (26% improvement) at 3 years
• PCI is currently recommended for patients with limited disease SCLC, who have complete response to systemic chemotherapy
• EORTC 08993 showed improved brain met control, improved DFS and OS in extensive stage disease patients with any response to chemotherapy
• RT dose 25/10 is standard (RTOG 0212) for limited stage, though 24-36 given in 2-3 Gy/day is used; RT dose 20/5 was used in extensive stage EORTC trial showing survival benefit
• Acute radiation toxicity is usually mild, and resolves within few months. Data on long-term neurotoxicity is lacking
• Concomittant chemotherapy can increase neurotoxicity, and should be avoided.
• PCI should start as soon after completion of induction chemo as feasible based on meta-analysis results

Meta-Analysis

• Brussels, 2001 (Belgium) PMID 11432756 -- "Prophylactic cranial irradiation in small cell lung cancer: a systematic review of the literature with meta-analysis." (Meert AP, BMC Cancer. 2001;1:5. Epub 2001 Jun 19.)
  • Meta-analysis. 12 randomized trials with 1547 patients, assigned to +/- PCI. (5 trials PCI initially with induction chemo, 2 trials PCI at end of induction chemo without checking response, 5 trials with CR only)
  • Outcome: decrease in brain mets incidence HR 0.48 (SS) for all pts +PCI, improvement in OS HR 0.82 (SS) only for +PCI pts in CR
  • Toxicity: long-term neurotoxicity not well described
  • Conclusion: PCI can be recommended for patients with SCLC and CR documented by brain scan

• Prophylactic Cranial Irradiation Overview Collaborative Group; 1999 PMID 10441603 -- "Prophylactic cranial irradiation for patients with small-cell lung cancer in complete remission. Prophylactic Cranial Irradiation Overview Collaborative Group." (Auperin A, N Engl J Med. 1999 Aug 12;341(7):476-84.)
  • Meta-analysis. Individual data of 987 patients from 7 randomized trials. Patients with complete remission. Extensive disease in 12-17%.
  • Outcome: 3-year OS PCI+ 21% vs. PCI- 15% (absolute benefit 5%, SS). 3-year LC 33% vs. 59% (SS). DFS also improved
  • RT dose: larger doses (8 Gy, 24-25 Gy, 30 Gy, 36-40 Gy) led to greater decrease in risk of mets, but no impact on survival
  • Timing: decreased risk of mets with earlier administration after induction chemo
  • Conclusion: PCI improves overall survival, DFS and control of brain metastases
  • Critique: 4/7 trials had <100 patients, ~14% had extensive disease, dose-fractionation not uniform

Randomized Trials

• EORTC 08993/22993 (2001-2006)
  • Randomized. 286 patients. Extensive stage SCLC and any response to 4-6 cycles of chemo. No brain imaging was required prior to randomization but rather screened for symptoms of brain mets. Treated with 1) PCI (20/5-30/12) vs. 2) no PCI.
  • 1-year; 2007 PMID 17699816 — "Prophylactic Cranial Irradiation in Extensive Small-Cell Lung Cancer." Slotman B et al. N Engl J Med. 2007 Aug 16;357(7):664-72.
    • 1-year outcome: decreased symptomatic brain mets: PCI+ 15% vs. PCI- 40% (SS), improved OS PCI+ 27% vs. PCI- 13% (SS); extracranial progression of 90%
    • Toxicity: PCI side effects, but clinically not significant
    • RT dose: 60% received 20/5 with impressive benefit. Given short survival, authors suggest shorter courses
    • Conclusion: PCI reduces risk of symptomatic brain mets, and improves DFS and OS in ED SCLC
    • Critique: Brain imaging not required prior to randomization or with followup. Therefore, patients may have had subclinical brain mets (that would have been seen on MRI) in the radiation group and extensive stage group. Applicability of this study to MRI negative extensive stage patients is difficult to interpret. NCCN has PCI with extensive stage with response to chemotherapy and good PS as category 1.
  • QoL; 2008 PMID 19047288 -- "Prophylactic Cranial Irradiation in Extensive Disease Small-Cell Lung Cancer: Short-Term Health-Related Quality of Life and Patient Reported Symptoms--Results of an International Phase III Randomized Controlled Trial by the EORTC Radiation Oncology and Lung Cancer Groups." (Slotman BJ, J Clin Oncol. 2008 Dec 1. [Epub ahead of print])
    • Planned QoL analysis (C30 questionnaire and Brain Cancer module). Compliance 94% at baseline, 60% at 6 weeks, 55% at 3 months. 3-month cut-off used for short-term HRQOL
    • Outcome: Mean global health status score higher in no PCI group at 6 weeks (SS) and 3 months (p=0.06). Mean score for hair loss and fatigue worse with PCI (SS), trend to worsening with other symptoms (e.g. appetite, constipation, N/V, social functioning, H/A, motor dysfunction, and leg weakness). Severe HRQOL worsening PCI 35% vs. no PCI 22% (SS).
    • Conclusion: PCI should be offered for survival benefit, but patients should be informed of potential adverse effects from PCI

• Gustave-Roussy PCI-88, 1998 (France) PMID 9857997 -- "Controlled clinical trial of prophylactic cranial irradiation for patients with small-cell lung cancer in complete remission." (Laplanche A, Lung Cancer. 1998 Sep;21(3):193-201.)
  • Randomized multi-institutional. 211 patients with SCLC in CR (goal 1100 patients). PCI protocol at discretion of each center (but RT 24-30 Gy, dose <=3 Gy, in <3 weeks). Mean F/U 5 years. Closed prematurely because results of PCI-85 convinced them to offer PCI to everyone.
  • 4-year OS: +PCI 22% vs. -PCI 16% (NS)
  • 4-year brain mets: +PCI 44% vs. -PCI 51% (NS)

• UKCCCR/EORTC, 1997 PMID 9470828 -- "Prophylactic cranial irradiation is indicated following complete response to induction therapy in small cell lung cancer: results of a multicentre randomised trial. United Kingdom Coordinating Committee for Cancer Research (UKCCCR) and the European Organization for Research and Treatment of Cancer (EORTC)" (Gregor A, Eur J Cancer. 1997 Oct;33(11):1752-8.)
  • Randomized. 314 patients. Limited stage SCLC with CR. Randomized to +PCI (36/18, 24/12, 30/10, 8/1, others as per treating MD) or -PCI
  • Brain mets: +PCI better with HR 0.44. DFS +PCI with HR 0.75. RT dose 24/12 same as -PCI, but higher doses better.
  • OS: no difference
  • Toxicity: Impairment in both groups before PCI, additional impairment at 6 months and 1 year, but no difference between +PCI and -PCI
  • Conclusion: PCI should be offered for CR. Dose should probably be >24/12

• RTOG Abstract, 1996 -- "A randomized phase III study of prophylactic cranial irradiation (PCI) vs. observation (OBS) in patients (Pts) with small cell lung cancer (SCLC) achieving a complete response: Final report of an incomplete trial by the Eastern Cooperative Oncology Group and Radiation Therapy Oncology Group." (Wagner Jr H, Proc Am Soc Clin Oncol 1996;15:876. (Abstract 1120).)
  • Randomized, stopped early. Median F/U 1.5 years
  • Brain mets: +PCI 19% vs. -PCI 53%; Median OS: +PCI 3.9 months vs. -PCI 6.5 months (NS)

• Gustave-Roussy PCI-85 (1985-1993) -- PCI 24/8 vs Observation
  • Randomized. 294 patients with SCLC, in CR after initial therapy. Limited (~80%) or extensive stage eligible. Arm 1) PCI 24/8 vs Arm 2) observation. Extensive neuropsych testing. Primary endpoint occurence of brain mets
  • 1995 PMID 7707405 -- "Prophylactic cranial irradiation for patients with small-cell lung cancer in complete remission." (Arriagada R, J Natl Cancer Inst. 1995 Feb 1;87(3):183-90.)
    • Outcome: 2-year isolated brain failure PCI+ 19% vs PCI- 45% 19% (SS); Development of brain mets 40% vs 67% (SS). 98% of brain mets in first 2 years. 2-year OS: PCI + 29% vs PCI- 21% (NS)
    • Toxicity: No difference on neuropsychological function or CT scan
    • Conclusion: PCI in patients with CR decreases risk of brain mets, potential effect on OS should be tested with higher power

• Okayama, 1993 (Japan) (1981-1986) PMID 8069603 -- "Comparative study of prophylactic cranial irradiation in patients with small cell lung cancer achieving a complete response: a long-term follow-up result." (Ohonoshi T, Lung Cancer. 1993 Oct;10(1-2):47-54.)
  • Randomized. 46 patients with CR, +/- PCI. Median F/U 8.5 years
  • Brain relapse: +PCI 22% vs. -PCI 52% (SS). Sole brain met: +PCI 4% vs. -PCI 17% (NS).
  • 5-year OS: +PCI 22% vs. -PCI 13% (NS)
  • Toxicity: late neurotoxicity infrequent; only 1 mild deterioration in PCI group

Other Reports

• Miami, 2001 PMID 11404503 -- "Twice-daily prophylactic cranial irradiation for patients with limited disease small-cell lung cancer with complete response to chemotherapy and consolidative radiotherapy: report of a single institutional phase II trial." (Wolfson AH, Am J Clin Oncol. 2001 Jun;24(3):290-5.)
  • Phase II. 27 patients with LD-SCLC and CR offered PCI 30-36 Gy in 1.5 Gy BID. 15 patients accepted, 12 patients declined. Median F/U 20 months
  • 2-year DFS: +PCI 54% vs. -PCI 0% (SS). 2-year OS: +PCI 62% vs. -PCI 23% (SS)
  • Toxicity: No difference

• 1973 PMID 4354042 -- "Should initial treatment of small cell carcinoma include systemic chemotherapy and brain irradiation?" (Hansen HH, Cancer Chemother Rep 3. 1973 Mar;4(2):239-41.)
  • Original suggestion for PCI

PCI dose regimens

• There is insufficient data to make a recommendation
  • Common fractionation scheme is 25/10 (this is also the "standard" arm in ongoing RTOG and EORTC trials)
  • There is a suggestion that 30-36 Gy in 2-3 Gy/fx or BED equivalent produces good response (although RTOG 02-12 suggesting no benefit to increased dose)
  • Doses below 24/12 are not effective
  • Results of RTOG/EORTC studies should provide clarification; in abstract publication (ASCO 2008) 25/10 appears comparable in brain control but with better overall survival over higher doses

• Intergroup PCI99-01 / EORTC 00223-08004 / RTOG 02-12 (1999-2005)
  • Randomized. 720 pts with L-SCLC, in CR after chemo+thoracic RT. Arm 1) PCI with standard dose 25/10 vs Arm 2) PCI with high dose (36/18 QD or 36/24 1.5 Gy BID). All had brain imaging at baseline. Primary endpoint incidence of brain mets at 2 years
  • 2009 PMID 19386548 -- "Standard-dose versus higher-dose prophylactic cranial irradiation (PCI) in patients with limited-stage small-cell lung cancer in complete remission after chemotherapy and thoracic radiotherapy (PCI 99-01, EORTC 22003-08004, RTOG 0212, and IFCT 99-01): a randomised clinical trial." (Le Pechoux C, Lancet Oncol. 2009 Apr 20. [Epub ahead of print]). Median F/U 39 months
    • Outcome: 2-year incidence of brain mets standard 29% vs. high dose 23% (NS). 2-year OS 42% vs. 37% (p=0.05)
    • Toxicity: fatigue 30/34%, headaches 24/28%, N/V 23/28%
    • Conclusion: No significant reduction of brain mets after higher dose PCI, but significant increase in mortality. PCI at 25/10 should remain standard of care

• Saskatchewan, 2003 (Canada) PMID 12788167 -- "Twenty-year follow-up study of long-term survival of limited-stage small-cell lung cancer and overview of prognostic and treatment factors." (Tai P, Int J Radiat Oncol Biol Phys. 2003 Jul 1;56(3):626-33.)
  • Retrospective. Registry review, 1417 SCLC, 244 limited stage.
  • 10-year cause-specific survival: 15%
  • PCI: no difference in dosing schedule (range from 15/10 to 33/12)

• Gliwice, 1998 (Poland) PMID 9531363 -- "Dose-response relationship for prophylactic cranial irradiation in small cell lung cancer." (Suwinski R, Int J Radiat Oncol Biol Phys. 1998 Mar 1;40(4):797-806.)
  • Meta-analysis. Analysis of brain relapse as a function of dose and time interval to PCI. "Early" PCI <60 days of primary treatment, "Late" PCI >60 days from treatment
  • Early PCI: no dose effect, no threshold dose. Late PCI: dose threshold ~15-20 Gy
  • Conclusion: For high probability of eliminating relapse, required dose 30-35 Gy in 2 Gy/fx

• Wisconsin, 1985 (1974-1984) PMID 4083270 -- "What is the lowest effective biologic dose for prophylactic cranial irradiation?" (Komaki R, Am J Clin Oncol. 1985 Dec;8(6):523-7.)
  • Retrospective. 327 patients treated with chemo (cyclophosphamide, doxorubicin, vincristine and/or MTX). PCI during first week after diagnosis, concurrent with chemo #1.
  • Probability of brain mets at 1 year: similar for 25/10 or 30/10, ~10%.
  • Toxicity: No clinically apparent late neurotoxicity

Neuro-toxicity

• There are concerns about neurotoxicity of PCI
• There is significant neurotoxicity present in patients completing induction chemotherapy, prior to PCI
• Data from randomized PCI trials is not well reported, but the neurotoxicity appears comparable with or without PCI
• Ongoing RTOG and EORTC trials will hopefully provide more information

• Mayo, 2004 PMID 15328816 -- "Cognitive dysfunction in patients with small-cell lung cancer: incidence, causes, and suggestions on management." (Kanard A, J Support Oncol. 2004 Mar-Apr;2(2):127-32; discussion 133-5, 138-40.)
  • Review. Cognitive dysfunction prevalence 15-90%

• Edinburgh, 1997 PMID 9470828 -- "Prophylactic cranial irradiation is indicated following complete response to induction therapy in small cell lung cancer: results of a multicentre randomised trial. United Kingdom Coordinating Committee for Cancer Research (UKCCCR) and the European Organization for Research and Treatment of Cancer (EORTC)" (Gregor A, Eur J Cancer. 1997 Oct;33(11):1752-8.)
  • See above for details. Randomized. 314 patients. Limited stage SCLC with CR. Randomized to +PCI (36/18, 24/12, 30/10, 8/1) or -PCI
  • Toxicity: Impairment in both groups before PCI, additional impairment at 6 months and 1 year, but no difference between +PCI and -PCI

• Gustave-Roussy PCI-85, 1995 (France) PMID 7707405 -- "Prophylactic cranial irradiation for patients with small-cell lung cancer in complete remission." (Arriagada R, J Natl Cancer Inst. 1995 Feb 1;87(3):183-90.)
  • See above for details. Randomized. 294 patients with SCLC in CR. Treated with PCI to 24/8 observation. Extensive neuropsych testing.
  • 2-year toxicity: no difference. Abnormality on CT scan comparable (+PCI 27% vs. -PCI 21%, NS)

• MD Anderson, 1995 PMID 7642416 -- "Evaluation of cognitive function in patients with limited small cell lung cancer prior to and shortly following prophylactic cranial irradiation." (Komaki R, Int J Radiat Oncol Biol Phys. 1995 Aug 30;33(1):179-82.)
  • Prospective. 30 patients with limited SCLC, CR to chemotherapy + thoracic RT/resection. Neuropsychological tests prior to PCI. 9 patients had prior neurologic history, 21 patients were "normal". 11 patients had post-PCI testing 6-20 months later
  • Initial eval: 29/30 (97%) had cognitive dysfunction prior to PCI. Mostly verbal memory, frontal lobe dysfunction, fine motor coordination. Of normal patients, 20/21 had impairment
  • Post-PCI eval: no significant difference
• Edinburgh, 1994 (UK) PMID 7654431 -- "Neurological and cognitive impairment in long-term survivors of small cell lung cancer." (Cull A, Eur J Cancer. 1994;30A(8):1067-74.)
  • Retrospective. 64 patients in remission >= 2 years, 52 with PCI, recalled for assessment
  • General function: 95% status <=1, 9/37 neuro exams abnormal
  • Neuropsychometric testing: 80% impaired on at least one exam, 54% impaired on two or more exams. Significant dysfunction. No conclusion about PCI, since few patients did not receive it
  • Dose-response: 8/1 less impaired than higher dose in multiple fractions

Decision Analysis

• MD Anderson, 2006 PMID 16877726 -- "Decision analysis for prophylactic cranial irradiation for patients with small-cell lung cancer." (Lee JJ, J Clin Oncol. 2006 Aug 1;24(22):3597-603.)
  • Evaluation of risk/benefits of PCI for limited stage SCLC (QALE = quality-adjusted life expectancy), by varying survival rates and neurotoxicity (NT). Neurotoxicity not well reported, so modeled "mild" or "moderate"
  • Baseline: 5-year OS +PCI 26% (from Turrisi PMID 9920950), 5-year OS -PCI 22% (estimated equivalent). If mild NT, QALE +PCI 4.31 vs. -PCI 3.7 (SS). If moderate NT, QALE +PCI 4.2 vs -PCI 3.7 (SS).
  • If hypothetically improved survival: 5-year OS +PCI 40%, -PCI 36%. +PCI still better for mild/moderate NT, but not worth it for severe NT
  • Conclusion: At current OS, with mild/moderate neurotoxicity, PCI is worthwhile. As OS improves, impact of neurotoxicity becomes more important, but PCI still worthwhile if mild neurotoxicity

To Be Categorized

• Abstract — "Is prophylactic cranial irradiation indicated in small-cell lung cancer?" Fleck et al. J Clin Oncol. 1990 Feb;8(2):209-14.
• Commentary. "Brain irradiation and systemic chemotherapy for small-cell lung cancer: Dangerous Liaisons?" Turrisi. J Clin Oncol. 1990 Feb;8(2):196-9.
• Mayo retrospective study. 45 Gy in 15 fractions, split course. 43 patients. All patients alive at 5 years had received PCI.
• Bonner, NCCTG. Abstract, not yet published. BID TRT with daily chemotherapy (etoposide and cisplatin), with PCI given in 3rd-4th week, and TRT started 4th-5th week. No difference in cognitive ability or quality of life. Local control no better than conventional treatment, but 5year OS 24%, MS 20 months.
• Abstract Full Text — "Phase II trial of up-front accelerated thoracic radiotherapy combined with chemotherapy and optional up-front prophylactic cranial irradiation in limited small-cell lung cancer. Groupe d'Oncologie Thoracique des Regions Alpines." Hugli et al. J Clin Oncol. 2000 Apr;18(8):1662-7.

  Swiss study. PCI given during 1st cycle.

• Meta Analysis, 17 trials. Rate of brain mets, 60% vs 40%. Slight survival advantage. Subgroup analysis showed no difference between dose fractionation schemes.