The role of intensity modulated radiotherapy in gynecological radiotherapy: Present and future

Abstract

Aim

This manuscript reviews the English language literature on the use of intensity modulated radiation therapy (IMRT) for gynecologic malignancies, focusing on the treatment cervical cancer.

Background

Radiation therapy plays a key role in both definitive and adjuvant treatment of these patients, although efforts continue to minimize acute and chronic toxicity. IMRT is an attractive option because of the potential to dose escalate to the target while sparing organs at risk.

Methods and Materials

The English language literature was reviewed for relevant studies.

Results

Multiple heterogeneous studies have showed dosimetric and clinical benefits with reduction in acute and late gastrointestinal, genitourinary and hematologic toxicity, especially in the post hysterectomy scenario and for dose escalation to para-aortic nodes. Consensus is evolving regarding necessary margins and target delineation in the context of organ movement and tumor shrinkage during the course of radiotherapy. Protocols with daily soft-tissue visualization are being investigated.

Conclusions

Consistency in approach and reporting are vital in order to acquire the data to justify the considerable increased expense of IMRT.

1. Introduction

Intensity modulated radiation therapy (IMRT) is an external beam modality which uses variable intensity across the face of the beam to shape isodoses to achieve a high tumor dose while minimizing exposure to healthy tissue. The combination of 3D planning and variable radiation intensity in each field provides dosimetric advantages which have been exploited in a variety of pathologic sites including cancer of the cervix.

Cervix cancer management varies depending on the FIGO stage, but radiotherapy plays a vital role across the range of presentations. For early stages, treatment may consist of surgery or radiotherapy alone, but in the presence of adverse prognostic factors, surgery will be combined with radiotherapy. For bulky or locally advanced presentations, combined radio-chemotherapy is the standard of care. Phase III randomized trials showing the benefit of concurrent cisplatinum with pelvic radiotherapy also provide toxicity data. Acute and chronic grade 3–4 gastrointestinal toxicity is 7–16% and genitourinary up to 17%.^1–6 These toxicity rates increase when radiation fields are extended to include the para-aortic regions, with grade 3–4 acute gastrointestinal toxicity in up to 49% and chronic toxicity seen in up to 34% at 36 months. Grade 3 or 4 hematologic toxicity is reported in 76%.^1,7,8 RTOG 0116 combined chemotherapy and extended field radiotherapy, and found acute nonhematologic grade 3–4 toxicity of 81%, and chronic grade 3–4 toxicity of 40% with follow up ranging to 38 months.⁷

Follow up is relatively short in many of these studies, and despite the already high toxicity, the situation may worsen with time. Two decades ago, Eifel et al. reported that the risk of serious complications from radiotherapy increases with time but at different rates depending on the organ system studied. Retrospective analysis was performed on 1784 patients with carcinoma of the cervix treated with radiation. Grade 3 toxicity levels at 3 and 5 years were 7.7 and 9.3% but increased approximately 0.34% per year through 10–20 years. Although most serious complications are diagnosed within 2–3 years, the risk continues to increase steadily up to 25 years after treatment, especially in the genitourinary system. This underlines the need for improvement in radiotherapy delivery.⁹

2. Dosimetric benefits of IMRT

Early work on IMRT showed advantages compared to 3D conformal radiotherapy in dose reduction to the organs at risk for radiation toxicity.^10–13,14 Roeske et al. compared the dose received by the small bowel, bladder and rectum in ten patients with gynecologic cancers treated with either 3D conformal or IMRT. The V100 of the small bowel was reduced by 50% (p = 0.0005) and the V100 of the rectum and bladder by 23% (p = 0.0002 and p = 0.0005 respectively).¹³ When IMRT is used to deliver a 20–30 Gy boost, Chan et al. found a significant reduction in high dose volumes in 12 patients with cancer of the cervix, vagina or endometrium compared to the use of 3D conformal or a 4 field box. The V66 of the rectum was reduced by 22% (p < 0.001) and the bladder by 19% (p < 0.001).¹⁵

Mell et al. reported the dose volume histograms for organs-at-risk for 7 patients with carcinoma of the cervix treated with chemotherapy concurrent with IMRT, 3D conformal, or an anterior-posterior parallel opposed pair and found that dose to the bone marrow and small bowel was reduced, but the reduction to the rectum and bladder was less impressive. Hematologic tolerance was improved with less grade 3–4 toxicity by reducing low-dose irradiation to the bone marrow; V20 was 99, 97.8 and 72% with AP-PA, 4-Field box, and Bone marrow sparing -IMRT.¹⁶ The question of how much dose reduction is required, and to what volume, has not been answered. Simpson et al. suggested that a decrease in the V45 of the small bowel by 100 cc reduced grade 2 toxicity by 50% ¹⁷. Mell et al. found a correlation between hematologic toxicity and the volume of pelvic bone marrow receiving 10–20 Gy.¹⁶ This has been confirmed by more recent studies.^18,19

3. Impact on toxicity

The dosimetric advantages of IMRT have resulted in reduction of both acute and chronic GI and GU toxicity.^{12,20,21–25} Early retrospective studies by Mundt et al.^12,23 showed a significant decrease in acute grade 2 GU toxicity from 91 to 60%, and chronic GI toxicity from 20 to 3% with the use of IMRT rather than a 4-field box. Efforts to demonstrate superiority of IMRT over 3D conformal have produced preliminary data from many small retrospective studies with short follow up and including a heterogeneous mixture of definitive radiotherapy and post-operative patients. Doses range from 45 to 60 Gy with either pelvic or extended fields, and boosts are a mixture of brachytherapy, IMRT, or an integrated IMRT boost. Table 1 summarizes the retrospective data on toxicity.²⁵

Table 1.

Retrospective studies on toxicity comparing IMRT to other RT techniques.

Author/year	n	Patients	Technique	Field	Planning target volume	Dose	Follow-up (median)	Acute toxicity	Chronic toxicity	p-Value
Mundt 200212	40	Cervix/Endoa	EBRT 4Fb vs IMRTc	Pelvic	CTVd + 1 cm	45 Gy	NS	GIe gr2: 60% GUf gr2 10% No G3	na	p = 0.002 p = 0.22 Reduced ToxgG2 for IMRT vs EBRT 4F
Mundt 200323	36	Cervix/Endo I-II 70% RT Adjuvant	EBRT 4F vs IMRT	Pelvic	CTV + 1 cm	45 Gy	19 m	N/A	GI 19% grade not specified	p = 0.001 Reduced Tox compared to 3Dch
Chen 200711	68	Cervix RT Adjuvant	EBRT4F vs IMRT+ CTi	Pelvic	na	50.4 Gy	14 m	GI 1–2 36% GU 1–2 30%	GI 6% GU 9%	p < 0.05, but Late Tox GU p = 0.231
Beriwal 200721	36	Cervix RT Adjuvant Ib2-IVa	IMRT+ HDR-BTj + CT	Pelvic + parak	CTV 0.5 + 1 cm	45 Gy	18 m	GI gr 3:3% GU gr 3:3%	GU 3% Grade ≥3 2y-actuarial 10%
Kidd 201025	452	Cervix Ia-IVb	PET-CT guided IMRT+ HDR vs RTC3D + HDR/LDRl + CT	PET positive para, pelvic + para	CTV +7 mm	50.4 Gy	52 m	NA	G3 GU or GI IMRT 6% 3Dc 17%	p = 0.0017
Chen 201120	109	Cervix IB2-IVA	IMRT + BT + CT	Pelvic	CTV + 0.7–1 cm	50.4–54 Gy	32 m	GI 3:3%	GI 4.5%
Hasselle 201122	111	Cervix I-IVA Radical and Adjuvant	IMRT + HDR − BT CT	Pelvic	CTV + 0.7–1 cm	45 Gy	27 m	Grade ≥3 2%	Late tox g >  = 3:7%
Zhang 201224	58	Cervix I-II Adjuvant	IMRT + QT	Pelvic + Para	CTV +1CM	50.4 Gy	34 m	GU/GI G3 tox 5%	RTOG toxicity G3 5.1%

^aEndometrium.

^bExternal beam radiotherapy 4 fields.

^cIntensity modulated radiation therapy.

^dClinical target volume.

^eGastrointestinal.

^fGenitourinary.

^gToxicity.

^hRadiotherapy conformal 3D.

ⁱChemotherapy.

^jHigh dose rate brachytherapy.

^kParaortic field.

^lLow dose rate brachytherapy.

Evidence suggests that IMRT can spare bone marrow^16,26 but given the large volume of hematopoietically active marrow in the pelvis and lower lumbar spine, specific planning constraints are required. Otherwise, IMRT fails to show a clear advantage over 3D conformal, with hematologic grade 3 toxicity of 28%²¹ for extended field and 24% for pelvic fields.²⁰

Regarding efficacy (Table 2), no randomized comparisons of IMRT to other radiotherapy techniques exist but local failure rates, and overall and disease free survival appear to be similar for IMRT compared to 3D conformal. Haselle et al., reported on 111 cervix cancer patients with a median follow up of 27 months, treated with surgery or IMRT with or without brachytherapy. Overall survival at 3 years was 78% and disease free survival 69%.²² Zhang et al. included only post surgical patients and consequently had only a 3.4% local regional recurrence but a 27% metastatic failure rate.²⁴ Overall survival at 3 years was 71% and disease free survival 66%. The most common site of failure was distant, an event that can only be reduced by improved systemic therapy.

Table 2.

Results for pelvic Irradiation with IMRT in terms of local failure, disease free survival, and overall survival.

Author/year	n	Patients	Technique	Field	Planning target volume	Dose	Follow-up (median)	Local failure	Distant failure
Chen 200711	68	Cervix Adjuvant	4Fb vs IMRTa	Pelvic	na	50.4 Gy	14 m	EBRT: 4F 6% IMRT: 7%
Beriwal 200721	36	Cervix Ib2-IVa	IMRT	Pelvic + parac	CTV 0.5 + 1 cm	45 Gy	18 m
Kidd 201025	452	Cervix Ia-IVb PET CT staging	IMRT or 3Dc + HDR/LDR	PET positive Para, pelvic+ para	CTVd +7 mm	50.4 Gy	52 m	IMRT 8.1% 3Dce 10.4% p value ns	IMRT 15.6% 3Dc 24.6% p value ns
Chen 201120	109	Cervix IB2-IVA	IMRT + BTf	Pelvic	CTV + 0.7–1 cm	50.4–54 Gy	32 m	13.8% 3 years	22%
Hasselle 201122	111	Cervix I-IVA Radical and Adjuvant	IMRT + HDR – BTg	Pelvic	CTV + 0.7–1 cm	45 Gy	27 m	14%@ 3 years	17% @ 3years
Zhang 1224	58	Cervix I-II Adjuvant	IMRT – Para	Pelvic + Para	CTV +1 cm	50.4 Gy	34 m	3.4%	27%

^aIntensity modulated radiation therapy.

^bExternal beam radiotherapy 4 fields.

^cParaortic field.

^dClinical target volume.

^e3D conformal.

^fBrachytherapy.

^gHigh dose rate brachytherapy.

4. IMRT planning

Traditional external beam radiotherapy is based on field limits determined by bony landmarks. Large treatment volumes include generous amounts of healthy tissue but margins of security are large, such that target motion or change in GTV during treatment are not issues. In the era of conformal treatment with steep dose gradients, definition of the GTV and CTV become crucial. Major uncertainties exist regarding IMRT for cervix cancer in determining the required margins, the acceptable degree of homogeneity and the appropriate dose limits for the organs at risk. Such contouring demands a thorough knowledge of radiologic anatomy. Current RTOG protocols using IMRT include a contouring atlas for pelvic volumes and guidelines for dosimetric constraints.

Lim et al. published treatment guidelines for the radical treatment of cervix cancer based on studies of postoperative patients to create a consensus for the CTV and PTV of the primary tumor and regional nodes. There was moderate agreement on the contours of the cervix, uterus, vagina and parametria, but determination of margins was difficult given that these structures are subject to movement, deformation and tumor regression during treatment. There was lack of agreement on the parametrial limits, the length of vagina to be included in the PTV, and whether or not to include the entire uterus. Individual variation amongst patients makes the dynamic unpredictable. Margins of 1.5–2 cm around the tumor CTV and 7 mm around the PTV were suggested, but only with daily soft tissue visualization for position verification. If bony landmarks are used for set-up, then margins must be more generous.²⁷

Several studies have shown the importance of cervical tumor regression during treatment, with a median volume reduction of 46% (range: 6–100%).^14,28 This suggests that IMRT should be replanned during the final third of treatment to take advantage of the shrinking GTV. However, the reduction in the CTV and PTV may be disproportionally small. Van de Bunt et al. imaged 14 patients with MRI during treatment and reported that when GTV reduction exceeded 30 cc, re-planning reduced the small bowel in the high dose volume significantly. Margins of 10–15 mm were acceptable.¹⁴ Similarly, Lim et al. confirmed that DVH's did not accurately reflect the actual dose received by the target and organs at risk due to reduction in the GTV of 48–96% and CTV of 8–77% in 20 patients imaged weekly with MRI. He compared 3 IMRT plans with different margins and concluded that a margin of 5 mm may be adequate provided daily soft tissue visualization was used for treatment set up.²⁷

In addition to tumor regression, cervical-utero movement is also an issue in IMRT delivery. Margins for this mobile target are at variance with those required for the relatively fixed nodal CTV. Taylor et al. imaged 33 patients with gynecologic cancer using MRI and confirmed that since the uterus is more mobile than the cervix, asymmetric margins may be more appropriate.²⁹ Gordon et al. looked at 3 different dynamic models for the utero-cervical unit and concluded that the uterus would be under-dosed if margins were chosen based on cervical movement.³⁰ The bladder, rectum and small bowel are also dynamic mobile structures that effect the position of the GTV.^31–33 Protocols requiring a full bladder and empty rectum can minimize utero-cervical movement; such requirements are included in recent RTOG protocols.³⁴ The development of individualized strategies is essential due to the large variation within and between fractions for each patient. New protocols are being tested based on imaging techniques that deal with the variation in these structures during treatment, such as MRI, bladder ultrasound and CT.^31–33

Aside from under-dosing the target because of organ movement, tumor regression and inaccurate contouring, there are other significant risks to take into consideration with IMRT. Due to the multiplicity of fields and delivery angles, the integral dose to the patient is much higher than with other techniques (Fig. 1). Studies such as MEII08 suggest that the volume of healthy tissue such as the small intestine, rectum and bladder receiving a very low dose of radiation was higher with IMRT.¹⁶ This may lead to development of radiation induced malignancy.^35–37

Fig. 1.

Comparison of nodal boost plans, with respect to integral dose to the patient and the volume of tissue receiving low level radiation. (a) Parallel opposed pair plan. (b) 4-FIELD plan. (c) IMRT plan. Note the extensive volume enclosed by the 30% isodose in plan c.

5. IMRT and its impact on resources

The complexity of IMRT results in greater consumption of resources compared to traditional radiation techniques. The time required for physicist and oncologist to complete contouring tasks and treatment planning is much greater due to the detail required in contouring and the number of fields employed. To minimize positioning errors, immobilization must be certain and precise, and therapists take longer to position patients. IMRT demands more expensive and sophisticated planning systems, quality control and maintenance protocols. Recent analyses of cost effectiveness comparing IMRT with 3D conformal for tumor sites such as prostate and head and neck have shown that the use of IMRT for prostate cancer in the United States increased the cost for each quality adjusted life year saved by $50,000 compared 3D conformal.^38,39 In Canada, the increased expense is $27,800 over alternative techniques. In Spain IMRT costs 5–6 times more than 3D conformal. The variation between countries depends on health care billing practices and the details of how IMRT is performed.

6. The current role of IMRT

Dose escalation to the pelvis and para-aortic region with conventional radiotherapy is prohibited by unacceptable toxicity.⁸ The dosimetric benefits of IMRT have been concentrated in delivering a boost to these regions or a central boost to patients not suitable for brachytherapy. Simultaneous boosts to the target or affected lymph nodes have shown good results and the altered fractionation of a simultaneous boost has the advantage of shortening treatment time and improving the therapeutic ratio.^15,40

At present the role and potential benefits of IMRT have been best established in the setting of post operative treatment thanks to the findings of RTOG 0418.³⁴ In 2008 Small et al. published guidelines for contouring post operative patients to establish consistency of reporting amongst institutions for ease of comparison of data.⁴¹ Preliminary results from individual institutions have shown a benefit for these guidelines for both early stage disease as well as for those with positive nodes.^11,24,42 The advantage of IMRT for sparing of bone marrow in patients treated for pelvic tumors are being achieved by identifying specifically the active marrow in the treatment field using PET, MRI or SPECT. Early results are promising.^43–45

7. Future directions

Improvements in toxicity allow evaluation of more aggressive treatment approaches. Dueñas Gonzalez has reported good results with more radical chemotherapy in locally advanced cervix cancer.⁴⁶

One of the more important advances in the last decade has been identification of genetic variability in the individual response to radiotherapy through genetic markers of radiation toxicity. Single nucleotide polymorphisms (SNP's) are variations in the gene sequence that effect a single base pair and appear in at least 1% of the population. There are 11.8 million SNP's registered in the NCBI genome data base. Numerous studies are focusing on phenotypes susceptible to cancer and to radiation toxicity. Phenotypic expression is complex and is influenced by many life style and environmental factors. There are not yet any consistent associations between certain SNP's and types of radiation toxicity, possibly because of the great number of treatment variables that are also known to influence toxicity, such as the type of radiation, dose, treatment volume, fractionation and use of concurrent chemotherapy. Research in this area is just beginning.⁴⁷

8. Conclusions

IMRT offers dosimetric advantages in organ sparing compared with traditional 2D or 3D radiotherapy in treatment of cancer of the cervix. This has maintained excellent long term cure rates while reducing GI, GU and hematologic toxicity, especially in patients treated with concurrent chemotherapy. IMRT has made possible safer dose escalation to the para-aortic region, or bulky GTV in patients nor suitable for brachytherapy. However, these results should be taken with some caution given the heterogeneity of the populations studied, the small numbers of patients, short follow up and great variability in margins, treatment fields and dose prescription. A multi center study has shown that the use of IMRT for postoperative cases is reliable, but a clear consensus is still lacking for the delivery of radical treatment. Recommendations for contouring of volumes and the required margins continue to be debated, given the inherent complexity of the pelvic dynamic, the movement and regression of the tumor and variable size and position of the neighboring organs. In addition to contouring and planning issues, daily treatment verification must include soft tissue imaging to ensure correct alignment on the target.

IMRT is a high complex modality with an infinite number of planning options regarding the number of fields and angles of incidence to cover the required volumes and to meet the dose restrictions for the organs at risk. The result is both a flexible and powerful tool but one which consumes vast resources in terms of physics, physicians, therapists and machine time. Further prospective multi center trials will provide the required data on disease control, toxicity and quality of life to fully integrate IMRT into the management of cervical cancer.

Conflict of interest

None declared.

Financial disclosure

None declared.