Esophageal Cancer - Management

Early Stage Resectable tumors (Tis, T1a, T1b upper esophagus N0)

• Surgical resection
• Post-operative RT is indicated for positive margins!
  

Resectable esophageal cancers (>T1bN0)

• Surgical resection
• Definitive chemoradiation (if non-surgical candidate)
• Pre-operative chemoradiation (40 Gy / 15 Fx)
  
• Pre-operative chemotherapy - controversial
  

You need to re-stage these patients with CT or PET-CT before surgery (assess response, rule out mets)

Lower GE junction cancers

• Peri-operative ECF chemotherapy (MAGIC Trial)
• Post-operative chemoradiation (MacDonald trial)
  

Palliative Esophageal Cancer

• Brachytherapy 20 Gy / 5 fractions (obstruction)
  
• External beam 30 Gy / 10 fractions (bleeding and obstruction)
  
• Stenting and/or dilatation (obstruction)
  
• Resection in very selected patients (bleeding or obstruction)
• Chemotherapy (obstruction)

----------------------------------------------------------------
Pre-operative Chemoradiation
There is a lot of conflicting data for pre-operative chemoradiation. Even meta-analyses show conflicting results in terms of overall survival. In general, most show small, but statistically small improvements in overall survival.

Two major Phase III studies compared chemoradiation to chemoradiation followed by surgery for Squamous cell esophageal cancers. Both studies failed to show a survival benefit for the addition of surgery.

Bundenne (FFCD 9102) treated 450 patients with 2 cycles cisplatin and 5FU and 46 Gy. Patients were then randomized to either surgery or an additional 3 cycles of chemo and 20 Gy radiation. There was no overall survival benefit (40 vs 34%) and a higher treatment-related mortality for the surgery group (9 vs 1%).

Stahl treated 200 patients with 3 cycles induction 5-FU, cisplatin, etoposide and leucovorin. Patients were then randomized to either >66 Gy radiation and cisplatin/etoposide or 40 Gy and cisplatin/etoposide followed by surgery. This study demonstrated better local control for the surgery group (65 vs 40%), but ultimately there was no difference in overall survival (~25% @ 3 years).

If you are going to go this route, a recent meta-analysis by Gebski showed a significant benefit at 2 years overall survival with an absolute improvement of 13% and a hazard ratio of 0.81. When analyzed for sub-type, adenocarcinomas benefit. The two studies above were negative, but included only squamous cell cancers. Basically, this is still investigational, but it is fair to say that this may be viable treatment option and these patients should be enrolled into any active protocols.

Pre-operative Chemotherapy
Similarly, there is conflicting data for pre-operative chemotherapy. Gebski's meta-analysis also looked at pre-op chemo. There is a 2 year absolute 7% overall survival benefit for adding chemotherapy pre-surgery for adenocarcinomas.

Definitive Chemoradiation
The RTOG 8501 trial compared 64 Gy alone to 50.4 Gy and concurrent 5FU/cisplatin x 4 cycles. This trial included 260 patients with T1-3N0-1M0 esophageal cancers. This trial demonstrated superiority of chemoradiation over radiation alone. There were improvements in 5 year overall survival 27% vs 0%.

A follow-up study INT0123 looked at radiation dose escalation with concurrent chemoradiation. Essentially it was cisplatin and 5 FU combined with either 64 Gy or 50.4 Gy. This study demonstrated a lower 2 year overall survival rate for the higher-dose arm (30% vs 40% each) and equivalent local relapse rates (~50%).


Lower GE Junction Esophageal Cancer

Peri-operative Chemotherapy
The MRC randomized 500 patients to either surgery alone or to surgery and peri-operative ECF chemotherapy (3 cycles pre + 3 cycles post) in the MAGIC trial. This study demonstrated an improvement in overall survival at 5 years (36 vs 23%). As a caveat, only 15% of these patients were GE junction tumors, most were gastric cancers.

Post-operative Chemoradiation
MacDonald's trial for gastric and gastro-esophageal junction cancers compared surgery alone vs post-op chemoradiation (45 Gy in 25 fx + 5FU/LV given 1 wk pre-RT, then during wk 1 & 5 of RT, and 2 more cycles). 3-year survival was 50% vs 40% favoring chemorads. Relapse rates were also better at 50 vs 30%. This is standard treatment for post-operative GE junction cancers.

Treatment Volumes as per RTOG 0436
GTV = gross tumor
CTV = 4 cm longitudinally and 1 cm radially around primary tumor and 1cm expansion around any nodes

• Cervical esophageal cancers (10-15 cm): include the supraclavicular lymph nodes
• Middle esophageal cancers (15-30 cm): paraesophageal nodes
  
• Distal esophagus cancers (> 30 cm) : include celiac nodes
  

PTV = CTV + 1 cm

Technique:
Plan 1: Use AP/PA fields for first up to 39.6 Gy
Plan 2: Use AP and 2 posterior obliques up to 50.4 Gy
Basically want to spare the spinal cord!

Cervical esophageal cancers (cover supraclavicular LNs):
RTOG0436 recommends:
0 to 39.6 Gy: AP/PA
39.6 to 50.4 Gy: 2 anterior obliques, 1 PA field + electron boost to cover supraclavicular nodes

Esophageal Cancer - Staging

Esophageal cancer patients tend to have locally advanced at time of presentation as there is no serosa covering the esophagus to act as a physical barrier. Intramural lymphatics allow for early access and spread of tumor to lymphatics even in tumors invading the lamina propria (T1a) or the submucosa (T1b) .

Squamous cell and adenocarcinoma are the two most common histologies in esophageal cancer. Adenocarcinomas represent 75% and SCCs represent ~25% of all esophageal cancers. SCC carries a worse prognosis. The risk factors are different for each entity.

Squamous cell carcinoma risk factors include: smoking, alcohol use, Achalasia (esophageal motility disorder), tylosis (hyperkeratinization of palms and soles), prior thoracic irradiation, prior head and neck cancer, and Plummer-Vinson syndrome (iron-deficiency anemia, glossitis, esophageal webs).

Adenocarcinoma risk factors include: smoking, gastric reflux, Barrett's esophagus, prior thoracic irradiation.

Other rare histologies include: lymphoma, sarcoma (leimyosarcomas most common), melanoma, neuroendocrine (small cell), adenoid cystic, mucoepidermoid carcinoma

Initial presenting symptoms are invariably dysphagia and weight loss. Other symptoms include cough, odynophagia and hemoptysis. Voice hoarseness suggests likely involvement of the left recurrent laryngeal nerve.

Initial work-up after full history and physical include:

• CBC, LFTs, SMA7, albumin, protein, Alk phos
  
• Endoscopic ultrasound + biopsy
• Panendoscopy
• CT chest and abdo
  
• PET scan is better than CT for assessing nodes and mets (Sensitivity ~90%)
• Barium swallow
• PFTs pre-RT

Anatomy
The Cervical Esophagus is found 15 to 20 cm from the incisors it is bounded by the hypopharynx and the sternal notch. The Upper Thoracic Esophagus is found from 20 - 25 cm bounded by the sternal notch and the Azygous vein. The Middle Thoracic Esophagus is at 25 - 30 cm bounded by the Azygous Vein and the Pulmonary Arteries. The Lower Thoracic Esophagus is at 30 - 40 cm bounded by the Pulmonary Arteries and the GE Junction.


TNM Staging
T1a = lamina propria
T1b = submucosa invasion
T2 = invades muscularis propria
T3 = invades advetitia (no serosa)
T4a = Resectable tumor invading pleura, pericardium, diaphragm
T4b = Unresectable tumor invading adjacent structures (ie trachea, aorta, vertebra)

N1 = 1-2 regional LN
N2 = 3-6 regional LN
N3 = > 6 regional LN

AJCC Staging (The 2009 7th ed AJCC stages SCC and Adenoca's differently and adds Grade and location as criteria.... really annoying)

SCC incorporates both grade and location (for stage I & II)
Stage IA: T1N0 G1
Stage IB: T1N0 G2-3; T2-3N0 G1 L
Stage IIA: T2-3N0 G2-3 L; T2-3N0 G1 U/M
Stage IIB: T2-3N0 G2-3 U/M; T1-2N1
Stage IIIA: T3N1; T4aN0; T1-2N2
Stage IIIB: T3N2
Stage IIIC: T4aN1-2; T4bNx; TxN3
Stage IV: TxNxM1

Adenocarcinomas only use grade for Stage I & II
Stage IA: T1N0 G1-2
Stage IB: T1N0 G3; T2N0 G1-2
Stage IIA: T2N0 G3
Stage IIB: T3N0; T1-2N1
Stage IIIA: T3N1; T1-2N2; T4aN0
Stage IIIB: T3N2
Stage IIIC: T4aN1-2; T4bNx; TxN3
Stage IV: TxNxM1

Justifications for using such a complicated staging system.... data analysis demonstrates that prognosis is affected by grade, location, and histological cancer type.

5-year OS
Stage I: 50-60%
Stage II: 40%
Stage III: 20%
Stage IV: <5%

Prognosis is the same between SCC and Adenocarcinoma because of the changes to the TNM staging system. This wasn't the case before the change.

Oropharngeal Cancer - Locally Advanced

Summary:
For locally advanced oropharyngeal cancers the primary treatment option is concurrent chemoradiation at 70 Gy in 35 fractions with cisplatin every 3 weeks as per the Adelstein trial.

If the patient can not tolerate chemotherapy than there are two reasonable options either altered fractionation as per the Fu RTOG 90-03 study or combining cetuximab with radiation as per the Bonner study.

Finally, induction taxotere, cisplatin, and 5-FU chemotherapy can be considered for locally advanced tumors provided the patient can tolerate the expected toxicities.

Conventional radiation alone is a poor choice given the number of options listed above as it is proven to be inferior for both survival and local control.

Surgery should generally be reserved for salvage as the associated morbidity is universally unacceptable.

____________________

Locally advanced oropharyngeal cancers include stage III and stage IV tumors
Stage III = T3N0, T1-3N1
Stage IVA = T4aN0, T4aN1, T1-4aN2
Stage IVB = T4bNx, TxN3

These tumors have a 5 year overall survival of 30-40%. Local recurrence remains high and radiation therapy alone is not sufficient for adequate tumor control.

Treatment options include:

1. Chemoradiation - 70 Gy in 35 fractions with cisplatin
2. Altered fractionation - concomitant boost or hyperfractionation
3. Radiation and cetuximab
4. Induction chemotherapy followed by radiation or chemoradiation
   
5. Conventional radiation alone
6. Surgery +/- post-operative chemotherapy and radiation as indicated

Chemoradiation:
Concurrent chemoradiation is the mainstay of treatment in most locally advanced head and neck tumors and oropharyngeal cancers is no exception. The MACH-NC meta-analysis of 93 randomized trials by Pignon of locally advanced head and neck trials have demonstrated a significant benefit of chemoradiation compared to radiation alone. The latest 2009 update shows an absolute improvement of 6.5% in overall survival when concurrent radiation is used.

The French GORTEC study is an oropharynx specific study consisting of stage III and IV tumors. It compared conventional radiation to concurrent carboplatin/5-FU and conventional radiation. This demonstrated an improvement in 5 year overall survival from 16% to 22% and an improvement in local control from 25% to 50%.

A standard regimen is based of the Adestein Intergroup trial. This 2-arm study compared conventional radiation to concurrent chemoradiation and to split-course concurrent chemoradiation in unresectable head and neck cancers. This study demonstrated worse toxicity for the chemoradiation arm, but an improved 3 year overall survival of 40% to 20% in favor of the chemoradiation arm. The split-course arm was worse compared to the concurrent arm and was most likely due to tumor repopulation during the treatment break. The concurrent radiation arm gives cisplatin every three weeks on days 1, 22, 43.

Altered Fractionation:
This includes either concomitant boost or hyperfractionation with the goal of shortening treatment time or dose escalation, respectively. The RTOG 90-03 study by Fu provides level I evidence to support the use of either concomitant boost or hyperfractionation. In this study of locally advanced head and neck cancers these two regimens were superior when compared to conventional fractionation and split course hyperfractionation. Overall survival was improved at 8 years from 30% to 35%. Local control was improved at 8 years from 40 to 50%.

An EORTC study by Horiot included Stage II and III oropharyngeal cancers. This study compared hyperfractionated treated 80.5 Gy using 1.15 Gy fractions BID to conventional 70 Gy in 35 fractions. This also demonstrated a trend for improvement in 5 year overall survival 3o% to 38% and a significant improvement in local control 40% to 60%.

Cetuximab and Radiation
Bonner compared conventional radiation alone to conventional radiation combined with cetuximab (250mg/m2 weekly). The 2009 update demonstrated a 5 year overall survival of 45% vs 35%. Local control rates are improved from 40 to 50%. The update also correlated cetuximab related rash with an improvement in survival (HR = 0.5).

Induction Chemotherapy
Induction chemotherapy is largely used to reduce tumor burden and to improve rates of organ preservation. The MACH-NC meta-analysis shows a small absolute benefit of 2% for overall survival when induction chemotherapy is added. This is an option for patients with locally advanced tumors who can tolerate induction chemotherapy in addition to definitive treatment.

The two principal studies looking at induction chemotherapy added taxotere to a cisplatin and 5-FU regimen. Vermorken's EORTC trial compared cisplatin 5-FU induction to taxotere, cisplatin and 5-FU for three cycles. This study followed the induction chemotherapy with conventional or altered fractionation radiotherapy. The results demonstrated an improvement in overall survival of 37% vs 24%% and improved response rates 68% vs 54%. The TPF regimen was less toxic likely due to the decreased doses of cisplatin and 5-FU.

A second study published in the same issue was Posner's study. This study was essentially the same as Vermorken's study in regards to the induction chemotherapy. This study used concurrent chemoradiation instead of radiation alone after the induction chemotherapy. This study showed 3 year overall survval of 60% vs 50% in favor of the docetaxel/taxotere arm. Local control was better as well at 50% vs 40%.

Volumes:
All cases should be treated by IMRT.
GTV = gross tumor
CTV70 = GTV + 2 cm
CTV64 = Adjacent lymph node levels
CTV 56 = Elective nodal irradiation

PTV = CTV + 0.5 mm

Salivary Gland Carcinomas

Summary:
These tumors regardless of stage should be managed surgically upfront.

Indications to add adjuvant radiation are:

• Grade 3 tumor
• Positive margins
• Positive lymph nodes
• Tumor > 4 cm
• Adenoid cystic histology

Indications to add ipsilateral neck irradiation include:

• Grade 3 tumor
• Tumor > 4 cm.
  

Non-surgical candidates can be treated with definitive radiation.

Remember to cover cranial nerves VII (parotid) or V3 (submandibular) up to the skull base for adenoid cystic histologies.

____________________


As a general rule of thumb, the smaller the salivary gland, the greater the chance that a tumor is malignant. Submental (80%) > submandibular (80%) > parotid (20%).

Parotid gland tumors are mostly benign adenomas. These tumors can be resected and observed without additional treatment. The deeper and the more posterior a tumor is located in the parotid, the more likely it is malignant.

As part of the work up of a salivary gland mass, the patient should have a full history and physical. Initial workup includes a CT neck and chest. A FNA biopsy sent to cytology (to improve diagnostic rate). An MRI to help delineate and soft tissue extension. General blood work including a CBC, RFTs, LFTs, TSH and electrolye panel.

Staging:
T1: < 2 cm
T2: 2 - 4 cm
T3: > 4 cm
T4a: Invades mandible, ear canal, facial nerve
T4b: Encases carotid, invades skull base or pterygoid plates

N1: Single LN <> 6 cm

Stage I = T1N0
Stage II = T2N0
Stage III = T3N0, T1-3N1
Stage IVa = T4aN0, T4aN1, T1-4aN2
Stage IVb = T4bNx, TxN3
Stage IVc = TxNxM1

Surgery
Malignant salivary gland tumors are resected with margins > 5 mm as initial management. A full neck dissection is not required if elective nodal irradiation is planned. Parotid tumors should have a level IIA neck dissection. For submandibular gland tumors, a neck dissection can be considered for high grade or large tumors.

Adjuvant Radiation
Adjuvant radiation is indicated for:

• High grade tumors
• Large primary tumors (T3 or T4)
• Positive margins
• Positive lymph nodes
  
• Adenoid cystic or Squamous cell histology

Elective nodal irradiation should be offered to:

• Tumors > 4cm (T3 or T4)
• High grade tumors (they harbor microscopic disease in 50%)
• Squamous cell histology (50% recurrence without XRT to ipsilateral neck).

Adenoid Cystic Carcinoma
UCSF reviewed their Adenoid cystic cases and found that T4 tumors, perineural invasion, major nerve involvement and omission of adjuvant radiation where all predictors of recurrence (this makes sense). Hence, all Adenoid cystic carcinomas should receive adjuant XRT. These tumors rarely recur locally or involve regional lymph nodes, so elective nodal irradiation should not be routinely offered. However, they have a tendency to spread perineurally along cranial nerves in 50% of cases. Treatment volumes should include the cranial nerves to the skull base. Cranial nerve VII should be included up to stylohyoid foramen in Parotid tumors. The lingual branch of cranial nerve V3 should be included up to it's entry in the skull base or Meckel's Cave (represented by the arrows in the figures below) in Submandibular tumors.



Radiation Dose

• Gross disease should be treated with 70 Gy in 35 fractions
• Positive margins or extracapsular extension should be treated with 66 Gy in 33 fractions
• Negative margin tumor bed should receive 60 Gy in 30 fractions
  
• Elective neck treatment should receive 50 Gy in 25 fractions.

Parotid LN Volumes include I - IV in node negative tumors. Level V should be added if a LN is positive. Submandibular LN volumes include IA, IB, II, III. Levels IV and V should be added if a LN is positive.

Ideally, patients should be treated with IMRT techniques. If this is not available, wedge pair techniques are adequate when the volume to treat has a depth less than 7 cm. Otherwise, you can consider a mixed photon/electron technique.

Complications:
Xerostomia, cranial nerve paresthesia

Prognosis:
5 year overall survival

• Stage I = 75%
• Stage II = 60%
• Stage III = 50%
• Stage IV = 30%
  

5 year local control
Mucoepidermoid = 80% with radiation and 40% without radiation
Adenoid cystic = 75% with radiation and 25% without radiation

Most recurrences will occur distantly and not locally.

Head And Neck Initial Evaluation

A full history and physical should be obtained from the patient. Important points on history include: smoking and alcohol habits, a history of chewing tobacco, betel nuts, paan or arecca. Presenting symptoms are varied and range by site. There may be a change in voice quality, dysphagia, odynophagia, referred otalgia, bleeding, ulcer, weight loss and a painless neck mass. It is also important to get an idea of the patient's performance status as this may give an indication of how well a patient may tolerate an aggressive treatment.

Physical exam is also guided on presenting symptoms, but should include a full general exam. A full neck exam of all lymphatic stations should be completed. Lymph nodes should be described in terms of size, mobility, and consistency. An oral exam includes examining the oral and buccal mucosa and evaluating the dentition. The pallatine fossas can be examined. Manual palpation of the base of tongue is required if this is a suspected site of tumor. Laryngoscopy should be performed to evaluate the nasopharynx, oropharnx and hypopharynx. Vocal cords, base of tongue and epiglottis can be visualized.

Initial work-up includes biopsy of either the primary mass or a palpable lymph node. A fine needle aspirate is adequate for diagnosis. Routine blood work includes a CBC, electrolytes, Ca, Phosphate and Mg, liver function, renal function, baseline TSH, PT/PTT/INR. HPV testing can be considered. A CT scan of the neck and chest should be part of the initial imaging. MRI of the neck will help stage the tumor and evaluate for soft tissue invasion. A PET-CT can be performed particularly if there are borderline suspicious lymph nodes, but is not indicated as a routine for all patients.

Finally, the patient may need evaluation by medical oncology, dentistry, audiology, gastroenterology (or radiology) for prophylactic PEG placement in patients receiving chemotherapy and a consult to GI for pan-endoscopy to rule out synchronous primaries.