Neo-adjuvant radiotherapy in rectal cancer

Table 1 Tumor node metastasis-7 classification (2010) with subclassification of stage T3

TNM	Extension to
Tis	Carcinoma in situ: intraepithelial or invasion of lamina propria
T1	Submucosa
T2	Muscularis propria
T3	Subserosa/perirectal tissue
T3a1	Less than 1 mm
T3b	1-5 mm
T3c	5-15 mm
T3d	15+ mm
T4	Perforation into visceral peritoneum (a) or invasion to other organs (b)
N1	1-3 regional nodes involved
N1a	1 lymph node
N1b	2-3 lymph nodes
N1c	Small deposits in the fat
N2	4 or more regional nodes involved
N2a	4-6 lymph nodes
N2b	7 or more lymph nodes
M1	Distant metastases
M1a	1 distant organ or set of lymph nodes
M1b	More than 1 organ or to the peritoneum

¹This subclassification is based upon an evaluation using magnetic resonance imaging prior to treatment decision is clinically valuable, and recommended in this review. It can be used also in the histopathological classification but is not validated and not incorporated in TNM version 7. TNM: Tumor node metastasis.

Table 2 Subgrouping of localized rectal cancer assessed by magnetic resonance imaging¹ and the recommended primary treatment

Favourable “good” group	Intermediate “bad” group	Advanced “ugly” group
Mid/upper rectum	Mid/upper rectum
T1-3bLow rectum T1-2, T3aN0mrf clear	T3c/dlow rectum also includes T3bT4 with peritoneal or vaginal involvementonlyN1/N2mrf clear	T3 mrf positiveT4 with overgrowth to prostate, seminal vesicles, base of urinary bladder, pelvic side walls or floor, sacrum positive lateral lymph nodes
5 yr LFR2 < 10%	5 yr LFR2 10%-20%	5 yr LFR2 20%-100%
5 yr DFR3 < 15%	5 yr DFR3 15%-60%	5 yr DFR 30%-80%
Primary surgery (TME)4	Preop 5 × 5 Gy with immediate surgery5	Preop CRT or 5 × 5 Gy with delayed surgery6

¹The algorithm (modified from[102] with permission from the publisher Informa) does not primarily address the risk of systemic disease, although this risk also increases with the presence of many of “the risk factors”; however, not necessarily parallel to the local failure rate (LFR). The algorithm is also “too simplified”, in that a other factors like size of the mesorectum, anterior or posterior location, extramural vascular invasion (EMVI+) are also relevant.

²Calculated in the group of patients planned for surgery, i.e., irrespective of the surgical outcome. The table are valid if the surgeon is an experienced rectal cancer surgeon and no pre-treatment is given.

³The 5-year risk of distant failure (DFR) is also given, although this risk is not well established. Risk factors detectable on magnetic resonance imaging for distant failure are N2 (versus N0 and N1), EMVI+, mrf+ and all T4 (a and b, see Table 1). These are also the risk factors used in the ongoing trial[88], where patients at high risk failing systemically are included.

⁴A local procedure is possible in a few patients [chiefly pT1, sm1 (+ 2), N0]. This group is in the text referred to as “very favourable”.

⁵Preoperative chemotherapy (CRT) is also a valid option according to international clinical guidelines[21].

⁶CRT means chemoradiotherapy to 50.4 Gy in 1.8 Gy fractions with 5-fluorouracil (capecitabine). 5 × 5 Gy with delayed surgery should be used only in patients not fit for CRT.

Table 3 Major randomized radiotherapy trials in primary rectal cancer¹

Study	Inclusion time	No of patients	Treatments			Radiation technique2	Increased postop death	Local recurrence			Increased survival	Comments
			Surgery alone	Preop (C)RT	Postop (C)RT			Surgery alone	Preop RT + surgery	Postop RT
Pre-TME era
MRC1[91]	1975-78	824	Yes	5 Gy × 1 2 Gy × 10		AP-PA	No	43%	45% 47%		No	Very low radiation dose, no benefit
EORTC[92]	1976-81	466	Yes	2,3 Gy × 15		AP-PA	No	28%	14%1		No	Decreased local recurrence risk
Bergen[93]	1976-85	169	Yes	1.75 Gy × 18		AP-PA	No	24%	17%		No	Marginally decreased local recurrence risk, comparably low dose
Stockholm I[94]	1980-87	849	Yes	5 Gy × 5		AP-PA	Yes	28%	14%2	-	No	Increased postop death (8% vs 2%), large target, suboptimal technique, decreased local recurrence risk. Increased risk late complications
Uppsala[95]	1980-85	471	-	5.1 Gy × 5	2 Gy × 30	3D-C on RT	No	-	13%1	22%	No	Preop 5 Gy × 5 is better than postop RT (60 Gy). Increased risk of late complications after postop RT
S:t Marks[96]	1980-84	395	Yes	5 Gy × 3		AP-PA	Yes	24%	17%		No	Increased postop death (9% vs 4%)
MRC2[97]	1981-89	279	Yes	2 Gy × 20		AP-PA	No	46%	36%1		No	Slightly reduced risk of local failure, tendency to improved survival (HR = 0.79, 95%CI: 0.6-1.04)
North-West[98]	1982-86	284	Yes	5 Gy × 4		3D-C on RT	No	41%	18%3		No	Decreased local recurrence risk, 10 x 10 cm beams
SRCT[53,55]	1987-90	1110	Yes	5 Gy × 5		3D-C on RT	No	27%	12%3	-	Yes	Decreased local recurrence risk, no increased acute toxicity, some late toxicity after 10-15 yr
Stockholm II[99]	1987-93	557	Yes	5 Gy × 5		3D-RT	Yes	25%	12%3	-	Yes	Overlaps to a large part SRCT, simplified radiation technique, tendency to increased postop mortality (4% vs 1%). Lower local recurrence risk, increased survival as in SRCT. Increased risk of late complications
Post-TME era
EORTC 22921[38]	1993-03	1011	-	RT CRT3	RTCRT	3D-C on RT	No		17% 9%2		No	2 × 2 design, chemotherapy in addition to RT gives fewer local recurrences as first event than RT alone irrespective of whether concomitant (9%) or postoperative (10%), or both (8%), increased toxicity, no increased survival
FFCD 9203[37]	1993-03	742	-	RT CRT		3D-C on RT	No		17% 8%1		No	Preop CRT results in fewer local recurrences than preop RT, increased toxicity, no survival difference
AIO-94[30,100]	1995-02	823	-	CRT	CRT	3D-C on RT	No		6%2	13%	No	Preop CRT is less toxic and gives fewer local recurrences than postop CRT, no difference in survival
TME[54,101]	1996-99	1861	Yes	5 Gy × 5		3D-C on RT	No	11%	5%3		No	No increased postop mortality. Decreased local recurrence risk even with TME, no improved survival, some risk of increased late complications after 5-10 yr
LARCS[39]	1998-03	207	-	RT CRT		3D-C on RT	No		33% 18%1		Yes	The only study in “ugly” rectal cancers, preop CRT gives better local control and better disease and cancer specific survival, tendency towards better survival (66% vs 53% after 5 yr). Increased acute and possibly late toxicity from CRT
MRC-CR07[31]	1998-05	1350	-	5 Gy × 5	CRT if CRM+	3D-C on RT	No		5%2	11%	Yes	Preop 5 Gy × 5 better than postop CRT if CRM+, marginally increased survival. No increase in late complications (3-5 yr)
Polish[33]	1999-02	312	-	5 Gy × 5 CRT		3D-C on RT	No		11% 16%		No	First study that shows less risk of acute toxicity from 5 × 5 compared with preop CRT, no difference in local recurrence and survival or late complications (3-5 yr)
TROG[34]	2001-06	326	-	5 Gy × 5 CRT		3D-C on RT	No		7% 4%		No	Same design as the Polish study, same results

¹Only large studies of relevance for present treatment recommendations are included. Patients with tumours considered to be resectable were included in all studies but one (LARCS). Staging has varied considerably over the years, but most included patients belonged to the intermediate group (bad) except in the LARCS study where most tumours were locally advanced (ugly).

²AP-PA, anterior posterior beams with no blocking, meaning high radiation doses to large normal tissue volumes, 3D-CRT, 3D-conformed radiotherapy, 3 or 4 beams with blocking of normal tissues that did not contain tumour cells. 3D-RT (in the Stockholm II study) means 4 beams but no blocking.

³CRT means chemoradiotherapy with 1.8-2 Gy daily to 45-50.4 Gy. RT means the same radiotherapy as in the CRT arm without chemotherapy. ^aP < 0.05, ^bP < 0.01, ^cP < 0.001. TME: Total (or partial) mesorectal excision; CRT: Chemotherapy.

Table 4 Main differences between and potential advantages of short-course and long-course preoperative radiotherapy in intermediate (bad) rectal cancers¹

	Short-course	Long-course
Total (physical) radiation dose	25 Gy	45-50.4 Gy
Fraction size/number of fractions	5 Gy/5	1.8-2 Gy/23-28
Radiation duration	1 wk	4.5-5.5 wk
BED2, acute effects	37.5 Gy	37.5-44.4 Gy
BED2, late effects	66.7	72-84 Gy
Overall treatment time	About 10 d	10-14 wk
Demands of radiation resources	Planning + 5 fractions	Planning + 23-28 fractions
Concomitant chemotherapy3	No	Yes
Acute toxicity	Minimal	More
Late toxicity	Present, considered limited in the “bad” group	Present, but not extensively studied. Anticipated to be higher than after short-course
Down-sizing/down-staging	No4	Yes5

¹In locally advanced (ugly) tumours, long-course CRT is the preferred option although short-course RT with a delay to surgery is an option if CRT is not tolerated because of high age or co-morbidity;

²Biologically effective dose according to the time-corrected linear quadratic model. Major uncertainties exist in the relative biological efficacy of the fractionation schedules concerning the acute, antitumour effects. The parameters selected for the acute effects were those used in the meta-analyses from 2001[7], even if they can be criticized and probably are incorrect. For late effects, an α/β of 3 Gy with no time correction is used. The anticipated antitumour effects do not thus differ substantially and late toxicity is at least not higher with short-course RT;

³Improved local control with long-course RT, increased acute toxicity and probably also late toxicity. Should not be given with short-course RT;

⁴Seen after short-course RT with delayed surgery;

⁵Not relevant in these intermediate tumours (unless organ-preservation is aimed at), however, relevant in locally advanced (ugly) tumours. BED: Biologically effective dose.