Hadron therapy

W. De Neve

Context • Generic name

– Hadrons: particles susceptible to strong (nuclear) forces (άδρός = strong) – Neutrons, pi-mesons, protons and other atomic nuclei

• Belgium – Radiotherapy using photons or electrons – Hadron therapy is not available

• Europe – 14 operational centers

• Protons: 12 • Carbon ions and protons: 2

– 18.000-40.000 €/patient • World

– ~ 50 centers (USA: 11; Japan: 9) – Cost/patient

• USA: ≥100.000 $ /patient • Japan: ~40.000 € /patient

• Very few Belgian patients are referred for hadron therapy

Dimitri Mendeleev’s periodic table of elements

Electrons Protons Anti-Protons

Helium Carbon Iron

Photon Proton Carbon ion

Intra-cellular distribution of ionizations

Physics rationale: Little in front, almost nothing behind

W. De Neve

10 mm

10 μ

W. De Neve

Protons or photons

Biological rationale for carbon ions: harder downstream in tumor

1 nm 10 mm

Relative biological effectiveness (RBE)

• RBE is the inverse ratio of the doses required for a biological response

• The standard of comparison is cobalt gamma-rays or megavoltage x-rays

RBE(exp) = D(cobalt)/D(exp)

1 10 100 LET

RBE as function of LET in normal tissues R

BE

LET-RBE: challenges for clinical use

• Proton therapy – Similar to x-ray therapy – Conversion factor ~1.1

• Carbon ion therapy – RBE varies

• Along the beam trajectory • With LET, fractionation • Between normal tissues • Between cancers

– Tolerance data: 10,000 patients, 2 centers

Photon RT

Neutron RT

Equal growth delay

RBE = D(neutron) D(photon)

Variety of tumors

Variety of RBE-values

Tumor RBE-values generally higher than the 3.0-3.5 value, measured for normal tissues

Batterman et al. Eur. J. Cancer 17: 539-548; 1981

Adenoid Cystic Carcinoma: RBE-values ≈ 8

Adenoid cystic carcinoma; local control in unresectable disease

C-ion boost

Photon

Schulz-Ertner D., et al. Cancer 2005; 104:338–44

p = 0.08

C-ion at NIRS

Hasegawa at NIRS-MedAustron meeting 2013: 186 ACC, 64 GyE/16 F, 75% LC

From biophysics to clinical use • Hadrontherapy

– 2 to 8 times less dose in normal tissues • Protontherapy: reducing toxicity

– Children • Preserving growth and development • Minimizing cancer induction

– Adults • Carbon ions: resistant cancers

– Mainly adults

Medulloblastoma

Proton

Photon

METHODS AND MATERIALS: We performed a retrospective cohort study of 558 patients treated with proton radiation from 1973 to 2001 at the Harvard Cyclotron in Cambridge, MA and 558 matched patients treated with photon therapy in the Surveillance, Epidemiology, and End Results (SEER) Program cancer registry. Patients were matched by age at radiation treatment, sex, year of treatment, cancer histology, and site. The main outcome measure was the incidence of second malignancies after radiation. RESULTS: The median duration of follow-up was 6.7 years (interquartile range, 7.4) and 6.0 years (interquartile range, 9.3) in the proton and photon cohorts, respectively. The median age at treatment was 59 years in each cohort. Second malignancies occurred in 29 proton patients (5.2%) and 42 photon patients (7.5%). After we adjusted for sex, age at treatment, primary site, and year of diagnosis, proton therapy was not associated with an increased risk of second malignancy (adjusted hazard ratio, 0.52 [95% confidence interval, 0.32-0.85]; P=.009).

Incidence of second malignancies in adult patients Matched-pair analysis proton therapy versus x-ray therapy

Chung CS1, Yock TI, Nelson K, Xu Y, Keating NL, Tarbell NJ. Incidence of second malignancies among patients treated with proton versus photon radiation. Int J Radiat Oncol Biol Phys. 2013 Sep 1;87(1):46-52.

Proton therapy practice

• Procedures similar to x-ray therapy – More frequent use of surgical aids

• Markers • Spacers

• Pediatrics – Protocols of international study groups – Chemotherapy in proton centers

• Adults – Fractionation as in x-ray therapy

Carbon ion therapy practice

• Very frequent use of surgical aids • Few pediatric indications • Hypofractionation (average of 12 fractions)

– NIRS, Chiba, Japan • 4 fractions/week; 1-16 fractions • Combinations with other modalities • No RCTs

– HIT, Heidelberg, Germany • Hybrid schedules with x-ray therapy • RCTs

Why is hadron therapy not the standard?

HIT Heidelberg

accelerator

HIT Heidelberg

• Operating: 240 days/year (4-5days/wk) – 6 hrs treatment, 2hrs dosimetry – Maintenance: May & Aug

• Dual ring synchrotrons. – Circumference: 63M

• Existing treatment facility (passive beam)

– Experimental room x2 – Treatment room x3 (fixed beam ports) – Simulation room x 1

• New treatment facility (active beam) – Treatment room x 3 (fixed ports x2, gantry x1) – Simulation room x2 – Preparation room x6

NIRS treatment facility

Pediatric indications Proton centre Carbon-ion centre

Non-resectable osteosarcoma 32 2

Standard indications in adults

Chordoma and chondrosarcoma Various sarcomas

Melanoma (aerodigestive) Adenoid cystic carcinoma

Paranasal Meningeoma

Low-grade glioma

10 24

65 24 71

8 24

8 21

Model indication level I

Major salivary gland tumors Locally recurrent rectal cancer

H&N cancer re-irradiation

50 82

156

382 195

(Proton+carbon-ion)-centre

p C

choices 255 106

81 103

545

Indications treatable using established technology

CancerPlan Action 30

Re-imbursed

NOT re-imbursed

Model indications in adults No. TNM classification Pathology Cancer entity

Type of hadron therapy

1 Digestive tract tumors

Pancreatic cancer Locally advanced C

2 Digestive tract tumors

Rectal cancer Locally recurrent C

3 Lung & pleural tumors

Non - small cell lung cancer (NSCLC)

Sta ge III Inoperable

p

4 Head & neck tumors Major salivary glands tumors other than

adenoid cystic carcinoma

Primary & recurrent R ?1 * or inoperable Perineural invasion

C

5 Head & neck tumors Any Re - irradiation

6 Lung & pleural tumors

Non - small cell lung cancer (NSCLC)

Stage I Medically

inoperable/refusal

7 Digestive tract tumors

Hepatocellular carcinoma

Primary & recurr ent (Child - Pugh grade A or B) Size of < 3 cm: adjacent

tovessels or bile ducts or the gastrointestinal tract

Size of > 3cm

Prostate and breast cancer: not studied

5 Adenoid cystic carcinoma of head & neck (all stages)

30 21 Carbon ions 21

6 Paranasal tumors (all stages) 110 65 Protons 65 7 Meningioma benign and malignant 572 24 Protons 24 8 Low-grade glioma (grade 1 & 2) 236 71 Protons 71

Carbon-ions Protons

61 162

Model indications 9 Pancreatic cancer (all stages) 1338 78 Locally advanced inoperable Carbon-ions 401

10 Rectal cancer (primary & recurrent) 2347 1956 Local recurrence Carbon-ions 82

11 NSCLC (stage III) 1229 588 Protons 588 12 Major salivary gland tumors other than

adenoid cystic carcinoma (aal stages) 107 50 Carbon-ions 50

13 Head & neck cancer (primary & recurrent) 2574 1560 Re-irradiation Protons 156

14 NSCLC (stage I) 1054 179 Carbon-ions 179 15 Hepatocellular carcinoma (all stages) 455 16

Primary & recurrent size <3 cm: adjacent to vessels or bile ducts or the

gastrointestinal tract; Primary & recurrent size >3 cm

Carbon-ions

364

Carbon-ions Protons

1076 744

TOTAL (all indications) Carbon-ions Protons

1137 906

Remarkable results in common cancers

NIRS

NIRS MDAH

Locally Recurrent Unresectable Rectal Cancer Chemo(x-ray)therapy

5-y survival 0-20% 5-y local control 0-50%

C-ion no previous irradiation (NIRS, n=136)

5-y survival: 45% 5-y local control: 93%

C-ion re-irradiation (NIRS, n=23) 3-y survival: 65% 3-y DS survival: 51%

Durante et al. Lancet Oncology (in-press)

Locally advanced pancreatic cancer

• 1/3 of pancreatic cancers at diagnosis • Unresectable due to local invasion • No distant metastasis • Standard treatment focuses on palliation • Autopsy: 1/3 free of M+ disease • n ≈ 400/y (B)

Pancreatic cancer Anatomical challenges

Biological challenges • Severe radioresistance of pancreatic cancer • High-LET trials before the CT-scan era

– Neutrons [1,2] – Negative pi-mesons [3,4]

• High-LET compared to photons – Beter palliative effects – Tumor disappearance at autopsy

• Normal tissues exposed to high-LET – Severe toxicity

1. Thomas FJ, et al. Am J Clin Oncol. 1989 Aug;12(4):283-9 2. Cohen L, et al. Cancer. 1985 Sep 15;56(6):1235-41 3. Bush SE, et al. Int J Radiat Oncol Biol Phys. 1982 Dec;8(12):2181-6 4. Kligerman MM,et al. J Can Assoc Radiol. 1980 Mar;31(1):13-8

Study No. of patients

Median survival (months)

Overall survival, (months) Toxicity

ECOG: Klaassen 1985 [1] Fluorouracil alone RT (40 Gy) + fluorouracil

44 47

8.2† 8.3†

21% (18) 11% (18)

Grade NR: 27% 51%

GITSG 1988 [2] SMF only RT (54 Gy) + fluorouracil and SMF

21 22

8*

10.5*

0% (18)*

18% (18) *

Grade ≥ 3: NR

50%

FFCD/SFRO: Chauffert 2008 [3] Gemcitabine alone RT (60 Gy) + fluorouracil + cisplatin

60 59

13* 8.6*

53% (12)* 32% (12)*

Grade 3-4: 40%‡

65.5%‡

ECOG: Loehrer 2011 [28] Gemcitabine alone RT (50.4 Gy) + gemcitabine

37 34

9.2*

11.1*

5% (24)*

12% (24)*

Grade 4-5: 9%

41%

1. Klaassen DJ, et al. J ClinOncol. 1985;3:373-378 2. Gastrointestinal Tumor Study Group. J Natl Cancer Inst. 1988 Jul 20;80(10):751-5. 3. Chauffert B, et al. Ann Oncol. 2008;19:1592-1599 4. Loehrer PJ Sr, et al. J Clin Oncol. 2011;29:4105-4112

Modern randomized controlled trials of photon radiochemotherapy for locally advanced unresectable pancreatic cancer

Abbreviations: ECOG: Eastern Cooperative Oncology Group; RT: photon radiotherapy; NR: not reported; GITSG: Gastrointestinal Tumor Study Group; SMF: streptozocin, mitomycin and fluorouracil chemotherapy; MMC: mitomycin-C; FFCD-SFRO: Federation Francophone de Cancerologie Digestive and Societe Francaise de Radiotherapie Oncologique. *statistically significant; †not statistically significant; ‡during induction phase of treatment.

Standard treatment: IMRT-gemcitabine Median survival: ≤ 12 months

High-grade toxicity: 20-50% of patients

GEM+CIRT for Locally Advanced Pancreatic Cancer Tadashi Kamada, NIRS, Japan

Total dose n 12mo 24mo

Local Control 43.2GyE 24 68% 28%

45.6GyE-50.4GyE 26 65% 65%

Overall Survival 43.2GyE 24 71% 21%

45.6GyE-50.4GyE 26 67% 50%

Overall Survival Local Control

0.4

G4 toxicity: 3 cases

Outcome comparison

• Chemo(radio)therapy – Median survival: 8.5-13 months – Severe toxicity: 30-50% G4-5

• C-ion+gemcitabine (NIRS high-dose) – Median survival: 24 months – Severe toxicity: 3/26 pts G4

• Concurrent gemcitabine at 1,000 mg/m2 • Single institute, small cohort

Locally advanced unresectable pancreatic cancer

Conclusions regarding C-ion + gemcitabine 1. Doubling of median survival 2. Reduced toxicity 3. Safety window for concurrent chemotherapy 4. External validity: neutron data 5. Cost-effective compared to chemotherapy alone

(± €10000/QALY or LYG) 6. Borderline cost-effective compared to gemcitabine

+ photon radiotherapy (± €21 000/QALY or LYG) 7. Uncertainties

Verwijzing naar buitenlands hadrontherapiecentrum

• E112/S2 document – Carbonionen: HIT Heidelberg, CNAO, Pavia – Protonen: Orsay Paris, PSI Villigen – Snelle en vlotte procedure

• 2014 nieuw KB, RIZIV initiatief – Stop E112/S2 procedure – Procedure met offerteaanvraag

RIZIV-procedure 14 stappen

Aanvraagprocedure 1. Rechthebbende krijgt MOC waarin hadrontherapie wordt voorgesteld en

contacteert via zijn behandelde arts een erkend radiotherapiecentrum, het “verwijzend centrum”.

2. Het verwijzend centrum stuurt een vraag naar het RIZIV, Dienst Geneeskundige Verzorging, DGV, om dossier te openen (via Hadrontherapie@riziv.fgov.be) en vraagt in parallel een advies tot behandeling aan het hadroncentrum. DGV, geeft een dossiernummer in afwachting van het volledige aanvraagdossier.

3. Het Hadroncentrum geeft zijn akkoord tot behandeling (of niet) aan het verwijzend centrum. Bij een negatieve antwoord wordt door het verwijzend centrum advies gevraag aan een andere hadroncentrum.

4. Het verwijzend centrum stuurt het volledige gemotiveerde aanvraagdossier en het positief advies tot behandeling in het hadroncentrum, met kostenramingen, naar DGV (per aangetekend schrijven).

5. DGV bezorgt het volledige aanvraagdossier binnen 5 werkdagen na datum van ontvangst aan de leden van de Akkoordraad ter advies (via “beveiligde site”).

Beslissing akkoordraad(1) 6. Binnen 5 dagen na ontvangst van het dossier bezorgen de leden van de Akkoordraad hun advies aan de DGV via mailbox

a. Een positieve beslissing in geval van unaniem positief advies (min. 2 adviezen per “bank”)

b. Een vraag om bijkomende inlichtingen, waarna de procedure per mail zich herhaalt

c. In geval van negatieve adviezen (te motiveren) zonder consensus, uitstel van beslissing tot na samenroeping in persoon van de akkoordraad op het volgende voorziene vast vergadermoment ( 2x per maand voorzien). Bij blijvende onenigheid beslist de voorzitter.

Opmerking : niet adviseren voor “eigen patiënt / rechthebbende”.

7. DGV deelt de definitieve beslissing binnen vijf werkdagen mee : - aan het verwijzend centrum die de beslissing met betaalgarantie aan het

hadroncentrum meedeelt - aan de rechthebbende - aan de VI van de rechthebbende

Indien de beslissing van de akkoordraad positief is worden de kosten van de behandeling en de eventuele transport- en verblijfskosten van de rechthebbende en zijn (eventuele) begeleider ten laste genomen (betaalgarantie – mogelijkheid tot voorschot betalen is voorzien).

• Minstens 2 leden van elk groep/”bank” moeten stemmen om de Akkoordraad toe te laten een geldige beslissing te nemen (1 bank = Hospitalen (univ+non-univ) en 1 bank = Verzekering Instellingen). De Riziv vertegenwoordigers stemmen niet.

• Elk negatief advies moet gemotiveerd worden. • Elk niet unaniem negatief advies geeft aanleiding tot een

tweede ronde om advies te vragen, eerst via 2de mail ( met synthese van de eerste adviezen). Indien geen consensus per mail, zal het dossier tijdens de eerstvolgende plenaire vergadering (2 X/maand – vast geprogrammeerd) besproken worden; in geval van blijvend meningsverschil neemt de Voorzitter (RIZIV) een beslissing.

• Er is geen beroepsprocedure bij de Akkoordraad voorzien.

Beslissing akkoordraad(2)

Facturen 8. Factuur voor het positief

behandelingsadvies en voor de behandeling door het hadroncentrum

9. Facturen van het hadroncentrum (zie punt 8) en factuur voor forfait coördinatie door het verwijzend centrum

10. Facturen van eventuele reis- en verblijfskosten van de rechthebbende en zijn begeleider, via verwijzend centrum naar het Riziv gestuurd

Betalingen

11. Betaling van de kosten van de behandeling* (*rekening houdende met een eventueel betaald voorschot)

12. Vergoeding van de reis- en verblijfskosten van de rechthebbende (en zijn begeleider)

13. Betaling van advies aan het behandelend hadroncentrum

14. Betaling van coördinatie aan het verwijzend centrum

Procedure E112/S2 (boven) versus RIZIV (onder)

Week 1

MOC H-Centrum accepteert

Week 2 Week 3

Voorbereiding H-therapie

Week 4

Start H-therapie E112/S2

Week 1 Week 2

Stappen 1-4 1. MOC 2. Dossiernummer 3. Akkoord H-Centrum 4. Aanvraagdossier/offerte

Week 3

Stap 5 Dossier

Akkoordraad ≤ 5d

Week 4(+x)

Stap 6 Advies

Akkoordraad ≤ 5d

Week ++

Stap 6b Week 5(+x)

Stap 7 Communicatie

advies Akkoordraad

≤ 5d

Week 6(+x)

Voorbereiding H-therapie

Week 7(+x)

Start H-therapie

Resultaat: enkel verliezers

• Patient • Verwijzende zorgverstrekkers • Leden Akkoordraad • Belastingbetaler

Conclusions

• Lowering unwanted radiation dose • Biological advantages of high-LET hadrons • Main issue is cost

– Equipment-size reduction – Improving patient throughput – Severe hypo-fractionation

• Radiotherapy of the future if cost issues can be resolved – Fractionation: weeks (x-rays) to day(s) (high-LET hadrons) – Excellent patient tolerance – Improved repeatability – Control of large and x-ray resistant tumors – Better integration with systemic therapy and surgery

• Miserable Belgian RIZIV/INAMI referral procedure

METHODS: HRQoL data were prospectively collected on PRT-treated patients aged 2-18 treated at Massachusetts General Hospital (MGH). Cross-sectional PedsQL data from XRT treated Lucile Packard Children's Hospital (LPCH) patients provided the comparison data. RESULTS: Parent proxy HRQoL scores were reported at 3years for the PRT cohort (PRT-C) and 2.9years (median) for the XRT cohort (XRT-C). The total core HRQoL score for the PRT-C, XRT-C, and normative population differed from one another and was 75.9, 65.4 and 80.9 respectively (p=0.002; p=0.024; p<0.001). The PRT-C scored 10.3 and 10.5 points higher than the XRT-C in the physical (PhSD) and psychosocial (PsSD) summary domains of the total core score (TCS, p=0.015; p=0.001). The PRT-C showed no difference in PhSD compared with the normative population, but scored 6.1 points less in the PsSD (p=0.003). Compared to healthy controls, the XRT-C scored lower in all domains (p<0.001). CONCLUSIONS: The HRQoL of pediatric brain tumor survivors treated with PRT compare favorably to those treated with XRT and similar to healthy controls in the PhSD.

Pediatric brain tumors treated at MGH, Boston

Yock T, et al. Quality of life outcomes in proton and photon treated pediatric brain tumor survivors. Radiother Oncol. 2014 Oct 7 [Epub ahead of print]