Reirradiation of Recurrent Nasopharyngeal Carcinoma Near the Spinal Cord
Mr B, 61, presents with a third recurrence of nasopharyngeal carcinoma 34 months after primary chemoradiotherapy. His original plan delivered 70 Gy in 35 fractions to PTV70 with a simultaneous integrated boost: PTV70 received 70 Gy, PTV63 received 63 Gy and PTV56 received 56 Gy. The spinal cord maximum was 44 Gy and brainstem maximum 52 Gy. He now has a 2.8 cm recurrent mass in the nasopharynx abutting the clivus and extending to the cavernous sinus; there is no distant metastasis. The multidisciplinary team proposes reirradiation with 60 Gy in 30 fractions using VMAT, with a 5 mm margin. The cord and brainstem are within 8 mm of the GTV. The trainee asks whether this is safe. The department uses alpha/beta = 3 Gy for late CNS toxicity.
1. Calculate the EQD2 for the original spinal cord and brainstem doses and explain what the calculation shows.
Read the worked answer
Reasoning
The EQD2 formula is:
EQD2 = D × (d + α/β) / (2 + α/β)
where D is total dose and d is dose per fraction. For the original plan, the cord and brainstem doses were given in 2 Gy fractions, so with α/β = 3 Gy the EQD2 equals the physical dose:
- Spinal cord: 44 Gy in 2 Gy fractions → EQD2 = 44 × (2 + 3)/(2 + 3) = 44 Gy.
- Brainstem: 52 Gy in 2 Gy fractions → EQD2 = 52 Gy.
This is a useful simplification, but it does not mean the original plan was safe in isolation: 44 Gy to the cord is close to conventional tolerance, and 52 Gy to the brainstem is also near accepted limits. The key point is that these are already substantial late-dose exposures. A second course adding, for example, 30 Gy in 2 Gy fractions to the cord would give a cumulative EQD2 of 74 Gy, which exceeds commonly used conservative cumulative cord limits. The calculation must be repeated for the actual reirradiation dose distribution, not just the prescription dose, because the cord may receive only a small volume dose.
For late CNS tissue, α/β is often taken as 2–3 Gy; using 2 Gy would give the same EQD2 for 2 Gy fractions but would penalize larger fraction sizes more. The trainee should also note that the original 70 Gy PTV dose is not the relevant cord dose; the cord maximum of 44 Gy is the starting point.
2. What radiobiological and clinical factors determine whether reirradiation is safe, and what cumulative constraints would you propose?
Read the worked answer
Reasoning
Reirradiation safety is not determined by one number. Important factors include:
- Interval: 34 months allows some late recovery, but CNS recovery is incomplete and poorly quantified. Longer intervals are generally safer.
- Volume and dose distribution: a small cord volume receiving a high dose is different from a long segment. Brainstem volume effects are also important.
- Fraction size: 2 Gy fractions are less damaging than large fractions; the proposed 60 Gy in 30 fractions is conventional fractionation.
- Cumulative dose: the original cord EQD2 of 44 Gy and brainstem EQD2 of 52 Gy must be added to the new plan's EQD2, not to the prescribed 60 Gy.
- Tumour factors: recurrent nasopharyngeal carcinoma is aggressive; underdosing may sacrifice control, but catastrophic CNS toxicity is unacceptable.
There is no universally accepted cumulative limit. A conservative practical approach is to keep the cumulative spinal cord maximum at or below about 60–70 Gy EQD2 (α/β = 2–3) and the brainstem at or below about 70–80 Gy EQD2, while recognizing uncertainty and individualizing after MDT review. Those are not guarantees, and the plan should aim as low as reasonably achievable. If the proposed 60 Gy plan gives a cord maximum of 30 Gy in 2 Gy fractions, cumulative cord EQD2 would be 74 Gy, above a conservative 70 Gy limit; the plan would need to reduce cord dose, alter fractionation, or choose another strategy. The brainstem cumulative dose would similarly need calculation.
3. How would you counsel the patient and what planning or alternative strategies would you recommend?
Read the worked answer
Reasoning
Mr B needs an honest discussion of uncertain benefit and serious risk. He has a third recurrence, no distant disease, and limited curative options. Reirradiation may offer local control but carries risks of myelopathy, brainstem necrosis, cranial neuropathy, trismus, mucosal necrosis, and carotid blowout.
Counselling points:
- Explain that the original treatment already used substantial cord and brainstem dose.
- State that reirradiation is technically possible but that cumulative tolerance is uncertain; the plan will aim to keep CNS doses as low as possible.
- Discuss alternatives: surgery if resectable, systemic therapy, immunotherapy, palliative radiotherapy, or best supportive care.
- Obtain informed consent with documentation of the uncertainty.
Planning strategies:
- Use VMAT with steep dose gradients and a strict priority on cord and brainstem constraints.
- Consider reducing the prescribed dose to 54–60 Gy or using a simultaneous integrated boost that treats only the gross disease to the higher dose.
- Use daily CBCT and an adaptive plan if anatomy changes.
- Consider a 2–3 mm margin rather than 5 mm if set-up and imaging allow, while ensuring target coverage.
- If cumulative cord EQD2 cannot be kept within the agreed limit, do not proceed with the full dose; discuss a palliative or systemic approach at MDT.
The trainee should understand that '60 Gy in 30 fractions' is not automatically safe just because it is conventional fractionation.