By far the most informative review of ICR is an article in Endodontic Topics 2004 by Heithersay. In this paper he describes the diagnosis, treatment modalities, and expected outcomes for this particular type of external resorption.

Invasive cervical resorption can be a highly destructive form of external resorption, often times seen following trauma, "excessive" orthodontic forces, or internal bleaching with superoxyl. Like all forms of resorption, several factors must be present in order for the lesion to occur:
1. blood supply; 2. a stimulus; 3. breakdown or loss of the protective layer. When referring to the protective layer, what I'm referring to is the pre-cementum externally, and the pre-dentin internally. It might seem odd that these seemingly frail tissues are able to protect tooth structure. The reason this protection occurs is that specific peptides named RGD peptides provide binding sites for the resorptive cells. These RGD peptides are bound to calcium salt crystals on mineralized surfaces. So when the resorbing cells reach the unmineralized pre-dentin or pre-cementum, they are unable to bind as no RGD peptides are present and thus, no resorption occurs.

So, it would seem that a lack of pre-cementum on the surface of a root may predispose to the development of ICR. This could happen due to trauma via a number of mechanisms such as scaling/root planing, internal bleaching, physical trauma or perhaps orthodontic movement of teeth. Dental trauma, internal bleaching and orthodontics have been associated with the development of ICR (Heithersay 1990). The condition also occurs without any pre-disposing factors, and it is suggested that a genetic lack of cementum (and therefore pre-cementum) may be present in those cases. In contrast with other forms of resorption however , pulpal pathology is not involved. What's more is that even in highly advanced forms, a normal pulp is often encountered during testing and treatment. This is very important in designing an adequate treatment plan for our patients. If endo alone is recommended, there will be little to no effect on the resorptive process, unlike with internal resorption for example.
Therefore, it is of the utmost importance that proper diagnosis is attained. For starters and in contrast to other forms of resorptions, there will usually be an asymptomatic tooth with a normal pulpal response. Radiographic examination is also very useful. Angled radiographs should be obtained to help differentiate from internal resorption, as well as aid in the location of the lesion. (*as a quick refresher, the "SLOB" rule is the best way to determine B-L location of the lesion, where CBCT is not available. These lesions are not always readily detectable with an explorer. CBCT affords us very unique diagnostic and treatment advantages over conventional radiography alone.) Often times, the canal is clearly visible within the ICR lesion as compared to internal resorption, which will appear contiguous with the canal space. Furthermore, a thorough history, normal pulpal response and cervical location of the lesion will help differentiate it from other forms of external resorption.
The figure above shows Heithersay's classification of ICR. Class 1 and 2 lesions have a very successful outcome in excess of 90-95%. Class 3 and 4 lesions would require such extensive soft tissue and osseous manipulation in order to afford the operator sufficient access that other treatment modalities should be considered.
Treatment usually consist of gaining access to the entire lesion and using trichloracetic acid (TCA) to attempt to destroy the resorptive tissue via coagulation necrosis. Many times we can attempt an internal repair via the modified endodontic access cavity. The difficulty comes from the fact that the resorption is not usually confined to a discrete area and can have multiple feeding channels from well below the gingival margin. The lesion can also be treated surgically, but in practice it can be difficult to access and may even require bone removal to gain direct vision of the lesion. Endo treatment becomes necessary in many cases in order to properly allow debridement and restoration of the lesion. In advanced cases, a miniscule amount of pre-dentin is all that separates the lesion from the underlying pulp.
Surgical repair can rapidly get destructive of both tooth structure and periodontal tissues which is why it is important to identify and treat (if indicated) the disease as early as possible.
CASE 1: external repair
Patient presents with 2 suspicious lesions on teeth #10,11. Patient was asymptomatic at time of evaluation. Clinical exam reveals that #10 responds WNL to cold, percussion, palpation, mobility, and bite. There is some BOP noted along the Palatal. The resorptive defect can clearly be explored subgingivally. Tooth #11 is responding slightly delayed to cold and is otherwise WNL. With the aid of CBCT, a diagnosis of ICR is made for both teeth. It is determined that due to the extensive nature of the defect for tooth #10, extraction and implant therapy would provide the most predictable prognosis. Tooth #11 however is the M abutment for a long span FPD. Loss of this tooth would result in a need for more involved replacement of edentulous teeth. Given the Grade 2 Heithersay classification, along with the significance of retaining this abutment, a treatment plan of RCT and immediate surgical repair of the lesion was consented on. Treatment consisted of one visit RCT and core deep resin core, followed by a full thickness limited Palatal flap for surgical repair and treatment with TCA. The defect is then restored with Geristore. Geristore has been shown to have the potential for re-attachment by long junctional epithelium when used sub-gingivally.
CASE 2: internal repair
Patient presented after an incidental finding during recall. There were no symptoms at any time. A diffuse lesion was noted at the cervical area on radiographic exam. Pulpal testing was WNL. CBCT shows a large diffuse ICR lesion with an entry point through the disto-cervical surface. Interproximal origin for ICR lesions present access issues for surgical correction so an internal repair was attempted. RCT was completed and the lesion was debrided and treated w TCA internally. Care was taken to identify the point of origin. Once the case was completed and restored with Geristore, a 15 blade was used to sever the sulcular tissue attachment at the level of the entry point.

